SUMMARY 4.

DIFFERENCE BETWEEN THE ACTUAL AND CALCULATED PERCENTAGE
STRENGTH OF PERFORATED PLATES,

(The difference is expressed as a gain when the actual value
exceeds the calculated and vice versa).

between the rivets is also useful in enabling engineers to correctly proportion rivetted joints so as to ensure the tensile strength of the plate, beween the rivets, being equal to the shearing strength of the rivets.

In calculating the stress per square inch of net area between the holes, the net area was regarded as equal to the gross area less the metal removed by parallel holes of the same diameter of the small ends of the actual holes. Some engineers deduct the mean diameter of the rivet holes.

This is, however, incorrect, as the injury by punching consists both of the metal removed by the taper of the holes, as well as the injury arising from the deterioration of the plate near them. The stresses per square inch of net section between the holes are as in summary 8.

In summary 6 are given the values of the ultimate stresses in terms

per cent of the unperforated plates from which the specimens were cut and the gains and losses compared with the solid plate are given in summary. 7.

Whilst in the drilled specimens there is a decided gain over the unperforated plate, it will be seen that in the lengthway specimens, 1, 1, 4 and 1-inch thick in which the holes are punched, there are losses respectively of 4.4, 16.6, 27.3, and 33.0 per cent, and in the punched crossway specimens losses of 12.5, 34.0, 24.4, and 34.1 per cent. Comparing the results of the punched and annealed, and punched and subsequently bored specimens, it appears that they are about equal, but both are inferior to the drilled plates. The losses and gains given above are calculated with reference to the net section of the plate between the rivet holes,

SUMMARY 6.

ULTIMATE STRESS PER SQUARE INCH OF NET SECTION OF PERFORATED PLATES COMPARED WITH UNPERFORATED PLATE UNPERFORATED PLATE = 100.

and they are consequently greater than those given in summary 4, in which the losses and gains are expressed in terms of the gross area of the perforated plates.

The particulars given in summary 7 show the actual effects upon the plates between the holes of the punching or drilling, but the values given in summary 4 give the losses and gains for the joint as a whole. Care should be taken not to confuse the two.

The manner in which the ductility of the perforated specimens is affected by the treatment they receive is shown, to some extent, by the elongation of the holes, as will be seen in summary 8.

Whilst in the,,,, and 1 inch drilled specimens, the elongations of the holes are, 25.0, 30.8, 33.9, and 34.8 per cent.; the elongations in the punched plates of the same thickness are only 10.4, 11.6, 6.1, 3.1,

and 1.9 per cent., thereby indicating the greater suddenness with which the punched specimens broke. The smaller elongation of the holes in the specimens bored after punching compared with those in which the holes were drilled, shows that the object of the boring after punching, that is, to remove the injurious effects of punching, was not fully realised, and that a larger zone metal should have been bored out in order to effect this.

A consideration of the appearance of the fractures of the perforated specimens often affords some information of the effect of the treatment to which the plates have been subjected. The full reports A. and B. show that when unperforated the fractures of all the plates, both thick and thin, were silky. This appearance is found to occur when steel plates break gradually, that is, after a definite amount of extension has taken

place; and this having occured invariably in the unperforated specimens indicates that it is the normal appearance, and that any deviation from it arises from some definite cause.

The particulars in the summary below are taken from Report D., E., F., G., and H.

Comparing the results in summaries 6 and 9, it is seen that in the cases where the ultimate stresses are equal to, or not much less, than the ultimate stress of the unperforated plate, and thus indicate that little or no injury has been caused by the perforation, a fracture silky, or in great part silky, in appearance was generally shown. Also, where there has been a marked loss of strength, then the fracture is granular or principally SO. A granular fracture is, however, not necessarily a sign of the plate having been injured, as it may arise from sudden breakage, or from the presence of a nick in the specimen, or other cause.