Planning of rehearsal of flex-rotative fatigue for rising of the curve SN

The form usually used to determine the curve SN, of materials used in making of mechanical components and structures, is through fatigue tests. Those tests many times are accomplished being used 5 (five) stress levels and 15 (fifteen)specimens in each level. That plan type supplies less necessary estimator when compared to the other experimental plans, for instance, the sol called great plan and commitment plan. To main reason of this disadvantage it is the fact of using the same number of specimens in all the levels. It is observed that less accurate results are obtained in low levels, once the failure happens with smaller frequency. For this reason, more specimens should to be submitted to tests in the lowest levels than in the highest. Increasing the number of specimens in the lowest levels, which have low probability of failure occurrence, it will be increase to total number of failures, what contributes to a better analysis of the data. This work has for objective to present an alternative experimental plan, to determine the curve SN 1, with objective of supplying more necessary estimator. It will be made a practical application for the planning of rehearsal of flex-rotative fatigue for rising of the curve SN in steel SAE 8620. (author)