Determination of Resistance to Abrasive Wear. XII. Comparison of Akron and Du Pont Abrasion Machines. Value of an Abrasive Index

Abstract

Abstract Abrasion tests on several sole and heel rubbers have been made on the Akron and du Pont machines to determine whether the same abrasive index (abrasion-resistance relative to a standard rubber) is given by both methods. The following conclusions are drawn from the results. (1) The two methods do not give the same abrasive index for the same rubber; a study of published data shows that this is true of at least some other abrasion test methods. In using an abrasive index to specify or control the quality of manufactured articles it may, therefore, be necessary to establish a separate index figure for each type of machine used. (2) Broadly speaking, the higher the abrasive index of a rubber by any one method, the higher is its indices by other methods, so abrasive indices can be converted approximately from one method to another by using a suitable factor or conversion formula; a formula for interconverting Akron and du Pont results is given. (3) Such conversions from one method to another are only approximate, however, because indices determine by different methods are by no means exactly related. The magnitude of the errors involved is indicated and discussed; in the present experiments the average discrepancy between an index measured directly and that calculated by the formula from the index given by the other machine, is about 20 per cent; corresponding figures from published data range from about 6 to 40 per cent. (4) Errors similar to those just mentioned, occur in predicting the relative service wear of different rubbers from laboratory abrasion tests. Although these errors seem large, the chances of a rubber that gives a satisfactory test result proving really poor in service are small. (5) The adoption of any system that prevents unsatisfactory material being put on the market is a valuable means of saving rubber. On the basis of the present results with the Akron and du Pont machines, it can be shown that by the application of a laboratory abrasive index test the resultant saving of rubber is at least two-thirds of what would be saved if acceptance or rejection could be based on actual service wear results—an ideal but impracticable procedure. (6) There is evidence that among rubbers of similar composition, e.g., sole and heel rubbers made by any one manufacturer, abrasive indices obtained by different methods are more closely related than among rubbers in general. For control and development work within the factory, therefore, the abrasive index system should prove more satisfactory than when applied to a varied assortment of rubbers of unknown composition. (7) Abrasion methods may differ in their power of discriminating between rubbers; thus, in the present experiments, the Akron test shows wider differences in abrasive index than the du Pont test, although the latter has the higher power of discrimination.