Skip to main content

Predictive through process microstructural modelling of rolling and extrusion of aluminium alloys for client oriented and flexible (VIRFAB)


Links between six of Europe�s top aluminium production companies and seven academic research institutions have been established or intensified through the VIR[FAB] Project, resulting in collaboration in several fields. This has lead to improvements in analytical and experimental techniques that have been utilized to aid the characterization of materials for the development of a through process model. The techniques reviewed within the VIR[FAB] project have been: plane strain compression, thermo power measurement, thermo power measurement combined with residual resisitivity measurement (He-temp) and metallographic characterization techniques. Through a process of round robins and close collaboration the effectiveness of the techniques and importance of variations in methodologies has enabled member partners to improve the accuracy and validity of their experimentation. This has resulted in higher quality data that is more representative of the material, and a greater degree of accuracy in the measurement of properties. These developments have lead to a better understanding of the influence of deformation parameters on the properties of a number of aluminium alloys and this new information has been used to aid the improvement of commercial processing and as input for the through process model.
A non-contact sensor capable of measuring aluminium surface temperature during hot rolling, extrusion, cold rolling and annealing, for all alloys, surface-textures and -conditions, with an accuracy better than 1% has been developed. The developed sensor is based on the principle of Auxiliary Reference Thermometry(ART), employing two reference radiators and two pyrometers. The sensor has been tested successfully for moving strip at temperatures between 350-475K, with an accuracy of 2K. Preliminary tests indicate that an extension of the measurement range up to 750K will be possible. The sensor functions independent of material composition, or microstructure and therefore may be applied at numerous positions of the aluminium rolling or extrusion process and can be used for all alloy systems with minimal recalibration. This is a distinct benefit over current pyrometry based detection systems that are available. The sensor, with a size of 200x400mm has to be positioned at a distance of 20mm above the strip; for emergencies and to protect the sensor from the head and tail of the strip during hot rolling a retraction device has to be incorporated in the positioning system. Outside of the aluminium industry the sensor can be used on other reflective surfaces to contact less measure temperature in the range 350 - 750K.
This review document benchmarks the state-of-the-art in terms of modelling microstructural evolution in wrought aluminium alloys, to provide a valuable resource for scientists and technologists, both university-based and within industry, regarding the relationships between processing, microstructure and properties, and to help map out the move towards through-process modelling capabilities. The document comprises three main sections. The first provides a set of six documents, which each describe the applications and processing-microstructure-property relationships for a range of aluminium alloy sheet and extruded products. The second comprises six substantial reviews which each gives an overview of the state-of-the-art on various aspects of microstructure and materials modelling relevant to the project. The third and final section gives a series of summaries highlighting key results from previous collaborative research projects relevant to the current work.
A standardized format is necessary to ease the exchange of simulation in- and output between partners modelling different phenomena of the same processing line. The definition of the format has proved difficult due to the scope of data that the finished file format must contain. The agreed format requires a conversion tool, allowing the necessary data to be extracted and read into the existing simulation software. This tool is easily programmed in standard languages by the users (e.g.: Fortran, C++). The quick and easy exchange of simulation data allows new cooperations between researchers possible in scale of VIR[*] and future research and development collaborations involving modelling.