For the satisfactory accomplishment of any mechanical project, it is essential to know the characteristics of the materials that will be applied in the manufacturing process, in such a way that the resulting tension should not cause failure. The mechanical properties of the materials are established by the use of experiments or laboratory tests, carefully programmed, that reproduce, as accurately as possible, the real working conditions. However, these mechanical tests are destructive. The destructive characteristic of the experiments can be a problem when the parts that need to be tested cannot be damaged. Another complexity is the determination of mechanical properties in specifi c points of the assembly, which is justifi ed mainly in foundry pieces where the cooling conditions, the microssegregations and the occurrence of defects are distributed in a heterogeneous way along the part. Moreover, since the determination of the chemical composition and the microstructure is performed using non-destructive tests, they should be able to substitute the mechanical experiments, if a relationship between them could be established. This work presents mathematical models generated through statistical techniques of multiple linear regression, for the determination of the mechanical properties of hypoeutectic and eutectic aluminium foundry alloys using no destructive analyses, microstructure and chemical composition.
Aluminum alloys; mechanical properties; mathematical model
