THE IMPACT TESTING OF METALS 



By F. C. THOMPSON, D.Met., B.Sc, 

 Lecturer in Metallurgy, University, Sheffield. 



The introduction of rapidly moving and severely stressed parts 

 in engineering design, especially, perhaps, in motor-car and 

 aeroplane contruction, led to a search on the part of engineers 

 for a test in which rapidly applied stresses were involved which 

 should supplement the ordinary tensile test where the rate of 

 loading is relatively slow. The work of Eaton Hodgkinson 

 (1847) and Wohler (1859) paved the way for the so-called 

 " fatigue " tests in which fracture of the sample is brought 

 about by the rapid repetition of stresses often considerably 

 below the elastic limit as determined in static tension. Many 

 types of such test, which for some time enjoyed considerable 

 popularity, have been devised, but in recent years, and especially 

 during the war, in connection with the inspection of material 

 for air-craft construction, the fashion has changed. The " im- 

 pact " test, in which fracture of the test-piece is effected by a 

 single sudden blow, has replaced the fatigue test, in some forms 

 of which millions of repetitions of comparatively small stresses 

 were required. 



The literature of the subject is now somewhat voluminous, 

 and an interesting bibliography is to be found in a paper by 

 Dr. W. H. Hatfield (Proc. Inst. Mech. Eng., 1919, p. 347). 



The earlier work on impact testing revealed most exasperating 

 discrepancies in the resistance of presumably the same material 

 to sudden and severe stresses. For several years a controversy 

 was carried on between those who claimed that a form of test 

 which yielded such discordant results on samples which, so far 

 as the usual methods of testing were concerned, were identical, 

 was not worthy of credence, and those who saw in it a method 

 so sensitive that variations shown by no other process were 

 revealed. The latter view has been shown beyond doubt to be 

 correct. Especially as a result of the work of Charpy and 

 Cornu-Thenard (Journ. Iron and Steel Inst., 191 7, ii, p. 61) 

 it has been demonstrated that, given really uniform material, 



264 



