Abstract
Considerable scientific study has been devoted to the tensile and other forms of mechanical test, but the bend test, though one of the most useful, has escaped any such treatment, with the consequence that it is, as often as not, carried out in a slipshod manner that leads to fortuitous results and gives little or no knowledge of the properties of the material. The paper correlates the ductility shown by tensile and bend specimens, and enables the ultimate performance of a bend specimen to be foretold with reasonable accuracy. It gives a large amount of data that may be found useful in preparing specifications.
Owing to the complexity of the problems involved in plastic bending, the subject does not lend itself to theoretical treatment, but by the use of empirical data very simple formulæ can be derived, sufficiently accurate for applying to commercial tests and correlating various test conditions with one another. The results given by the more usual forms of bend test are analysed, and the special characteristics of the tests are described. Several methods of carrying out tests are shown to be faulty and to lead to unnecessary straining of the metal; consequently widely different results are obtained by different operators working to the same specification. Suggestions are given for designing the most suitable apparatus and for modifying methods of test, to improve results and to overcome inconsistencies.
The special problems involved in making a bend test on a welded joint are also discussed, and recommendations for tests are made. The ductility developed by such a test is also correlated with the ductility shown by a tensile test on the weld metal itself.
Get full access to this article
View all access options for this article.
