Abstract
Due to economic and environmental issues, the components weight and so the materials thickness have been reduced. This thickness reduction, combined with thermal shocks undergone by the exchangers and by the brazing assembly, led to oligocyclic fatigue problems on exchangers components, and particularly on tubes. To improve the Valeo exchangers reliability, SN & EN fatigue curves have been performed to standardize in fatigue the thin materials used for the thermal exchanger manufacturing. The temperature influence has been taken into account for the SN curves, and a first version of a criterion has been formulated.
Several series materials of different thickness coming from different suppliers have been sampled. For each material, strips have been cut in the material reels. The half of the strips has been brazed according to a Valeo specific protocol and the other part not. All the strips have been machining by electro-erosion on tensile, SN and EN test-pieces.
The Wöhler curves (SN test) have been performed during constant stress amplitude with a sinus signal, and at different temperatures (R=0, 1; f=30-50Hz; T= -30, 20, 120°C).
The Manson-Coffin curves (EN test) have been performed with an extensometer during constant strain amplitude with a triangle signal(R=0,1; f= 0.2-5Hz; T=Room temperature).
The ultimate tensile strength has been determined for all the sample, the Not Brazed tubes materials presented a Rm roughly 40 % higher than the Brazed ones.
The Wöhler curves revealed that the Not Brazed materials had a better endurance in fatigue than the Brazed ones. However, the curves comparison with one of our suppliers (Z) showed that the background (brazing conditions, machine, and operator) strongly influenced SN fatigue tests.
The Manson-Coffin curves, which represent the oligocyclic domain, give us the parameters we could use from the Manson-Coffin equation: Total strain, fatigue ductility coefficient and fatigue strength coefficient. In order to remove the brazing influence it is more relevant to base the SN standard criterion on materials before brazing and to set it in Rm and cycles number for a given application. As there is no significant differences between the EN curves, no criterion has been formulated on EN test. Only SN criterion will be used for material validation.
KEYWORDS – Fatigue, aluminium, standard creation, heat-exchanger