The number of published experiments on the ultimate compressive resistance of mild steel columns is, notwithstanding the extent to which this material has been employed in this direction, not so complete or extensive as those upon wrought-iron columns, and a series of further experiments on full-sized built-up columns of various sections is yet a desideratum.

As an example of such tests, the following may be quoted 3: -

1 Lanza, "Applied Mechanics."

2 Transactions of the American Society of Civil Engineers, vol. xiii., August, 1884.

3Ibid., vol. xxi.

Two columns of the type shown in Fig. 163, but with the lower flange plate replaced by open latticing, were tested to destruction. One column was flat-ended at one end and pin-ended at the other, with a total length of 15 feet from centre of pin to flat end of column. The section consisted of one flange plate 11" X 5/16", two side plates 10" x ¼", and four angles 2¼" x 2¼" x ¼". This column failed by flexure and local buckling at 20'9 tons per square inch.

Fig. 133.

Fig. 134.

Fig. 135.

The second column was of similar type, but of larger scantlings, flat-ended at both ends, 24 feet 1½ inch in length, and consisted of one flange plate 13½" X 5/16", two side plates 12" x ¼" two angles 2½" X 2½" X 5/16", and two angles 2½" X 3" X 5/16".

This column also failed by flexure or local buckling at 19.3 tons per square inch. The average ultimate tensile strength of the angles and plates was 35.3 tons per square inch, with 22 per cent, extension in 8 inches. The steel, therefore, was of somewhat harder quality than that generally referred to in Chapter I (Mild Steel: Its Manufacture, Physical And Chemical Qualities).

From the representation of the practical results obtained from the testing machine, we may now pass to the consideration of the breaking weights of columns as proposed by various authorities who have approached the subject either from a theoretical standpoint, or who have proposed formulae more or less empirical to embody the ascertained results of experiment.

In Figs. 136 and 137 are plotted the values given by Professor T. Claxton Fidler for fixed-ended and round-ended columns of hard steel, mild steel, wrought iron, and cast iron. For the details of the mathematical treatment of the subject, the student is referred to the original works of that author.1

In Fig. 138 are plotted the values for the breaking weights of wrought-iron columns, both flat-ended and pin-ended, of various sections, as given in the formulas proposed by an American authority. Similarly the values of the breaking weights of mild steel, wrought-iron, and cast-iron columns, flat- and pin- or round-ended, as given by the formulas proposed by another American authority, are plotted in Fig. 139.

We may now proceed to the discussion of various practical sections of struts and columns of rolled mild steel, such as are usually found in ordinary construction.

Figs. 140 to 175 give typical sections of struts and columns commencing from the simplest and most elementary forms, and exhibiting the growth or evolution of the more complex sections produced by riveted combinations of simple forms. It is to be remembered that the figures are type sections only. The proportions of the various members and their relative sectional areas and positions will be determined in every case by the conditions to be met, and the amount of load to be carried, whether considered as a direct simple vertical load, or a combination, as is often the case, of vertical and transverse loading. The designer will endeavour naturally in cases of direct vertical load to equalize as far as possible the radii of gyration about all axes, while disposing his metal at the greatest possible distance from the neutral axis; but it will frequently happen that this is not feasible, and, in fact, is only theoretically met by the adoption of perfectly circular sections, such as the hollow circular cast-iron column or welded tube. In all cases of unequal moments of inertia about different axes and unequal radii of gyration, the proportions of - must be determined with reference to the least radius of gyration or the direction in which failure by flexure may be expected.

1 Professor T. Claxton Fidler,"A Practical Treatise on Bridge Construction." See also Proc. Inst. Civil Engineers, vol. lxxxvi.

Fig. 136.

Fig. 137.

Fig. 138.

Fig. 139.