The Use of Statistical Control in Corrosion and Contact Resistance Studies

Publisher Website
Google Scholar
Abstract
It is pointed out that valid predictions from experimental data cannot be made unless the data are in statistical control. This fact is a serious handicap in studies of certain corrosion processes where the magnitude of the effect being studied is of the same order as that of the variance. Contact resistance data are used as illustrative examples. Statistical criteria are described which can be used to aid in detecting and eliminating causes of lack of statistical control, and it is demonstrated how an experimental error in the contact resistance measuring apparatus was eliminated using these criteria. The apparatus is then used to study the effect of oxidation of brass on its contact resistance behavior.

This publication has 0 references indexed in Scilit: