7 
ON IMPACT TESTS. 1 
Impact Tests.—Report of the Committee, consisting of Professor 
W. H. Warren (Chairman), Mr. J. Vicars (Secretary), Professor 
Payne, and Mr. E. H. Santrer, appointed to consider and report 
on the Standardisation of Impact Tests. 
Impact Tests of Materials. 
Iv is considered that ordinary static tests, such as the tension test on 
standard bars or the cold bending test, do not reveal the capacity of the 
material to resist shocks. In rails, axles, tyres and drawbars, armour 
plates, ordnance, moving parts in engines and machinery, the stresses 
are all more or less suddenly applied. It has long been recognised that 
the ordinary static tests in such cases should be supplemented by impact 
tests. Again, Mr. C. Fremont has recently shown that impact tests on 
wire and wire ropes reveal the weakness and want of homogeneity in 
the material in a much more satisfactory manner than static tests. Numer 
ous examples could be quoted where normal static tests have failed to 
express or define in a satisfactory manner the resistance of material to 
suddenly applied loads. The ordinary tension test may give a reasonable 
strength per square inch, and a reasonable percentage of elongation ; 
but the same material, when subject to impact, may break without much 
deformation and reveal a coarse fracture. Impact tests reveal the brittle- 
ness of the material or its tendency to fail by suddenly applied loads. 
Resilience is a function which is the reverse of brittleness, and expresses 
definitely the resistance to impact. It represents two factors, one of 
which is elongation, and the other the stresses producing them. A metal 
that shows a fair resilience will always give a fair elongation in a state of 
tension, but the converse is not true, as will be shown by the results of 
tension tests and impact tests of a certain steel used for the manufacture 
of high explosive shells. 
The tensile strength per square inch of a steel bar tested in the ordinary 
way is not a true expression of the actual strength, or resistance to break- 
ing expressed as ‘tenacity,’ because, at the moment of fracture, the area 
contracts locally, and it is this contracted area which breaks. Again, 
this contracted area does not vary with the strength exactly, but depends 
also upon other qualities of the material. 
The hardness number, as determined by the penetration of a steel 
ball ora steel cone—represents more exactly the actual strength or tenacity 
of the material. The Brinell hardness number denoted by A is related 
to the ordinary tensile strength per square inch denoted by o thus :— 
pS Oa 
where C is approximately constant for steels of the same kind; thus, 
for structural steel having a tensile strength of 26-7 tons per square inch, 
A = 1439 and C =0°'186. 
The relation between impact and hardness tests has not been exten- 
sively studied, but it will be shown that there is an approximate ratio 
somewhat similar to that observed in impact and tension tests. 
In regard to experiments made by impact machines on plain and 
nicked pieces, in every case the effect of nicking or notching a bar is to 
bring out more decidedly the properties which we have termed brittleness, or 
want of resilience, and it is this test we propose to more especially consider. 
