of almost constant quality and strength. Taking this as the standard, let us see how the timbers chiefly used in shipbuilding compare with iron as to their tensile strengths in proportion to their weights. One feature in which all timbers differ from iron is in their want of uniformity of quality and tensile strength. Even when the utmost care has been taken to season timbers, considerable variations are found to exist, not merely in different logs, but in the strengths of different pieces cut from various parts of the same tree. Such causes as the existence of knots, cross-grain, &c. affect the strength; and it is very different lengthwise of the grain from what it is across the grain. Hence arises a difficulty in ascertaining the average strengths of timber materials, and one which is not easily surmountable; with the greatest care in the conduct of experiments, different investigators have reached very diverse results. Taking the best of these experiments, the following are the results for a few of the timbers most commonly used:-* 1 Doubtful values; Mr. Laslett gives 5700 lbs. as the upper limit for teak, 7400 lbs. for Dantzic oak, and 6700 lbs. for elm. These figures are based upon the experiments of Barlow, Tredgold, Hodgkinson, and others, of which an excellent summary is contained in the late Professor Rankine's works, as well as upon the more recent and valuable experi ments recorded in Timber and Timber Trees, by Mr. Laslett, Admiralty Inspector of Timber. Sir W. Fairbairn's tables have also been examined in comparison with the others. British oak may fairly be taken as the standard timber, and its weight per cubic foot is about one-ninth that of iron, while its ultimate tensile strength might be about one-fifth that of iron. Here, then, the timber apparently gains upon the iron in its ultimate strength compared with its weight; but it is easy to see that it does not really compare so favourably. First, the builder would have no certainty that any piece of oak he might select would reach the average of strength: it might fall so low as to be only oneeighth the ultimate strength of iron, some specimens tested having had that ultimate tensile strength. Second, to guard against possible defects not discoverable on the surface, and to meet the different range of elasticity, a larger factor of safety would be employed with the timber than with iron -about 10 for timber, as against 4 or 5 for iron. As a simple illustration, take the case of a tie-bar of oak, say, 1 square foot in sectional area; it would probably have an ultimate tensile strength of about 570 tons, but would only be trusted with a moving load of about 55 to 60 tons. An iron bar of equal weight would have a sectional area of square foot, and a tensile strength of 320 tons; but, owing to its superior elasticity and the confidence felt in its uniformity of strength, it would be trusted with a load of from 65 to 80 tons. Or, to state the comparison somewhat differently, an iron bar capable of safely sustaining the same load as the oak bar need only have an ultimate tensile strength of, say, 260 tons, which would be equivalent to a sectional area of 13 square inches. The oak bar would weigh 54 lbs. per foot of length; the equivalent bar of iron would weigh about 45 lbs. per foot of length. The same considerations apply to other timbers, oak being superior to most, if not to all of them: and in these considerations we find one of the explanations of the superiority of iron to wood in the combination of lightness with strength. Professor Rankine proposed 5 tons per square inch as the average ultimate tensile strength of ship building timber; but, in view of the more recent and extensive experiments which have been quoted, this estimate appears too high, and 3 tons per square inch would be sufficient allowance; 48 lbs. per cubic foot is about the average weight of these timbers. Their ultimate resistances to compression also require consideration, in comparison with the resistance of wrought iron to direct compression.* Here authorities differ widely as to the strength of wrought iron. Professor Rankine gives from 27,000 to 36,000 lbs. per square inch; whereas Sir W. Fairbairn fixed it at 70,000 lbs., on the authority of Rondelet, the tensile strength being 45,000 to 50,000 lbs. per square inch. If the mean of the two statements is taken, it will be found that the ultimate resistance of iron to compressive strains is very nearly the same as its resistance to tensile strains, and this is probably very near the truth. Taking the same timbers as in the list previously given, it appears from experiment that their ultimate resistances to compression are as follows: A fair average value of the compressive strengths of timbers used in shipbuilding, therefore, appears to be about 3 tons per square inch, which nearly agrees with Professor Rankine's estimate. Against these strains, moreover, the use of so *The iron is not supposed to fail by "buckling." See remarks on this subject at page 299. large a factor of safety as against tensile strains scarcely appears necessary. Supposing a factor of safety of 8 to be taken instead of 10, the safe working load, on an average, for timber subject to compressive strains would be about three-eighths of a ton per square inch: for wrought iron, the working load would be from 2 to 4 tons-say, 3 tons as a safe average. As regards compressive strains, therefore, timber in single pieces compares better with iron, in strength relatively to weight, than it does in resistance to tensile strains. All pieces in a ship, however, are liable to both classes of strains, and consequently wood is inferior to iron, its inferiority becoming more marked when one passes from single pieces to a combination. These factors of safety for both tensile and compressive strains have been determined chiefly from the practice of civil engineers, and are adapted to the conditions of fixed. structures which have to bear the working loads frequently. There is an important difference between such structures and ships; for the latter have to resist the maximum strains (described in Chapter IX.) only on rare occasions, and probably at long intervals, the strains ordinarily experienced being much less severe. It will be evident that a severe strain only occasionally applied is not so likely to produce serious damage as a less strain frequently applied, especially when the character and intensity of the latter strain are continually and rapidly changing, provided that the maximum strain does not surpass the limits of elasticity of the materials. For these reasons, shipbuilders do not restrict themselves to the factors of safety approved by civil engineers. At present there are no recognised factors for the different classes of ships, but the subject is receiving attention, and from the analyses of the conditions of strain in numerous successful and unsuccessful ships there will probably be deduced, ere long, useful rules for practice corresponding to those of the civil engineer. The moduli of elasticity of the two materials afford, perhaps, the readiest means of comparing their relative resistances to both tensile and compressive strains. Professor Rankine gave the following values: More recent experiments made in the Royal Dockyards on some of these timbers give somewhat different moduli of elasticity. English and Dantzic oak, for example, had moduli of about 1,900,000-greater than those assigned by Professor Rankine; whereas teak had a modulus of about 1,300,000, or little more than one-half that in the above list. On the whole, however, it seems not unreasonable to accept the average modulus proposed by Professor Rankine, viz. that timber shall be considered to have about one-sixteenth the modulus of iron. and wood act together, therefore, this is the should govern their equivalent sectional areas.* of weights per cubic foot, it will be remembered, is about 1 for wood to 10 for iron. No further remarks will be needed in illustration of the superior combination of lightness with both tensile and compressive strength, in single pieces of iron as compared with single pieces even of the best timber. When iron ratio which The ratio The resistance offered by a combination of pieces of timber to compressive strains does not compare less favourably with that of iron than does the resistance of a single piece of timber to that of a single piece of iron, provided * See the remarks on page 315. |