Statistical Tolerancing as it Relates to Quality Control and the Designer (6 times 2.5 = 9)

Three methods of tolerancing are described. In widest use is the method of extremes, which produces unnecessarily close limits on the components. For the same over-all limits on the assembly the probability method allows wider limits on the components, but there are some unrealistic assumptions involved. A third method is described which results in tolerances falling between the other two. A case history is discussed comparing the three methods.