-----+----------------+----------------+
3/8 or 0.375 | 3/8 or 0.375 | 31,740 | 28,000 |
3/8 or 0.375 | 3/8 or 0.375 | 31,380 | 26,950 |
5/8 or 0.625 | 1/4 or 0.25 | 18,820 | 18,000 |
5/8 or 0.625 | 1/4 or 0.25 | 18,750 | 17,590 |
5/8 or 0.625 | 3/16 or 0.1875 | 14,590 | 13,230 |
5/8 or 0.625 | 3/16 or 0.1875 | 15,420 | 13,750 |
5/8 or 0.625 | 1/8 or 0.125 | 10,670 | 9,320 |
5/8 or 0.625 | 1/8 or 0.125 | 11,730 | 9,580 |
---------------------------------------------------------------------+

It will be seen from the above that the punched bars had the greatest
strength, indicating that punching had the effect of strengthening
instead of weakening the metal. These experiments have given results
just the reverse of similar experiments made on boiler plates; but the
material, such as above experimented upon, is what should be placed in
boilers, tough and ductile, and the manner of, and care taken in,
punching contribute to these results.

It is usual to have the rivet holes one-sixteenth of an inch in
diameter larger than the rivets, in order to allow for their expansion
when hot; it is evident, however, that the difference between the
diameters of the rivet hole and of the rivet should vary with the size
of the rivet.

The hole in the die is made larger than the punch; for ordinary work
the proportion of their respective diameters varies from 1:1.5 to 1:2.

As I have before stated, the best plate joint is that in which the
strength of the plate and the resistance of the rivet to shearing are
equal to each other.

In boilers as commercially made and sold the difference in quality of
the plates and rivets, together with the great uncertainty as to the
exact effect of punching the plates, have, so far, prevented anything
like the determination either by calculation or experiment of what
might be accepted as the best proportions of riveted joints.

In regard to steel plates for boilers Mr. F.W. Webb, of Crewe,
England, chief engineer of the London and Northwestern Railway, has
made over 10,000 tests of steel plates, but had only two plates fail
in actual work; these failures he thought were attributable solely to
the want of care on the part of the men who worked the plates up.

All their rivet holes for boilers were punched in a Jacqua