172 SECTIONAL ADDRESSES. 
throws, but of the micro-structure of the metal itself. The first question 
which rises in our mind is why the cranks deyelop flaws at all. It is, of 
course, known that with ordinary structures one is able to calculate the 
stresses in them, but this is not so with a locomotive crank-axle. Not 
only is it being subjected to the stresses set up by revolying it while 
it is loaded with the weight of a portion of the locomotive on its axle- 
bearings and by the steam pressure on the pistons transmitted to the 
crank-pins, but it has to withstand the shocks set up by its running on 
the rails, which cannot be calculated. These include the pressure set 
up on the edge of the wheels when entering a curve at a speed other 
than that which the super-elevation is allowed for, running over uneven 
rail joints and crossings, and also what I believe is one of the worst, if 
infrequent, the striking of check rails. These stresses and the resultants 
of them are most severe at the corners of the crank-pins and at the radii 
where the webs or throws join the rounded portions of the axle. These 
are the points at which flaws usually occur. 
For about twenty years we have endeavoured to get the knowledge 
we have obtained into an ordered state, from observation and discussion 
with the metallurgists attached to the various manufacturing firms. 
Certain points are obvious, such as the necessity of a good micro-struc- 
ture, and whilst the details in connection with exactly what micro- 
structure is the best are somewhat uncertain and open to debate, we 
can with confidence say that the steel ‘ shall be as free as possible from 
non-metallic enclosures, and that the micro-structure should show 
uniformly distributed pearlite in a sorbitic or very finely granular or 
lamellar condition and be free from any nodular or balled-up cementite. 
It must also be free from any signs of segregation and from any coarse 
or overheated structure.’ (Extract from Midland Railway specification 
for crank-axle forgings.) The dimensions I haye given of the size of 
the block of metal from which the axle is made show that it cannot have 
received much work, and therefore any non-metallic enclosures present 
will be only slightly drawn out, and will not occur as threads as they 
do in bars of small diameter and even in steel tyres. One of the first 
observations we deduced was that the life of the crank in miles had a 
direct relation to the ductility of the test-bar taken across the section 
of the throw and near the centre of the original ingot. This is the point 
at which non-metallic enclosures are most likely to be found, as well 
as that at which the greatest stress occurs. The inference is obvious 
that a flaw is likely to develop at some sharp corner of such an enclosure, 
In a section of steel such as that which must be used non-metallic 
enclosures are very likely to occur, and so steps had to be taken to 
ascertain what the best practical remedy was. With decreased carbon 
content greater ductility was likely to follow, and this has been shown 
to be the case. In a word, it is toughness rather than strength which is 
required, and the studied consideration of these points has led to an 
increased life in miles of the crank-axles of the 3,000 locomotives owned 
by the Company, in spite of the fact that they have been constantly 
growing in size, in pressure on the pistons, and in the work expected 
from them, This is shown in the following curves, which represent 
the mileage of crank-axles scrapped in the last twelve years. 
