This section is from the book "Spons' Mechanics' Own Book: A Manual For Handicraftsmen And Amateurs", by Edward Spon. Also available from Amazon: Spons' Mechanics' Own Book.
By the term "conversion" is understood the cutting up of the log or balk timber to dimensions suitable for use, allowance being made for alterations in form due to atmospheric influence, even on well-seasoned wood. "While wood is in the living state, a constant passage of sap keeps the whole interior moist and the fibres distended, more especially towards the outside. When the tree is felled and exposed to the air, the internal moisture evaporates gradually, causing a shrinkage and collapse of the fibres according to certain laws, being always greatest in a direction parallel with the medullary rays. In straight-grained woods the changes of length caused by atmospheric effects are slight, but those in width and depth are great, especially in new timber. Ordinary alternations of weather produce expansion and contraction in width in wood of average dryness to the following extent: - fir: 1/360 to 1/75, mean 1/124; oak: 1/412 to 1/80, mean 1/140. A practical allowance for shrinkage in 9-in. deals is 1/4 in. for northern pine and 1/8 in. for white.
The subject of shrinkage in timber has been well dealt with by Dr. Anderson, in a Cantor Lecture at the Society of Arts. His observations may be summarized as follows. If Fig. 220 be taken as representing the section of a newly-felled tree, it will be seen that the wood is solid throughout, and on comparing Fig. 221 with this the result of the seasoning will be apparent. The action is exaggerated in the diagrams in order to render it more conspicuous. As the moisture evaporates, the bundles of woody fibres shrink and draw closer together; but this contraction cannot take place radially, without crushing or tearing the hard plates forming the medullary rays, which are unaffected in size by the seasoning. These plates are generally sufficiently strong to resist the crushing action, and the contraction is therefore compelled to take place in the opposite direction, i.e. circumferentially; the strain finding relief by splitting the timber in radial lines, allowing the medullary rays in each partially severed portion to approach each other in the same direction as the ribs of a lady's fan when closing.
The illustration of a closing fan affords the best example of the principle of shrinking during seasoning, every portion of the wood practically retaining its original distance from the centre. If the tree were sawn down the middle, the cut surfaces, although flat at first, would in time become rounded, as in Fig. 222; the outer portion shrinking more than that nearer the heart on account of the greater mass of woody fibre it contains, and the larger amount of moisture. If cut into quarters, each portion would present a similar result, as shown in Fig. 223. Figs. 224-228 show the same principle applied to sawn timber of various forms, the peculiarities of which are perhaps indicated more clearly in Fig. 230. If we assume the tree to be cut into planks, as shown in Fig. 229, it will be found, after allowing due time for seasoning, that the planks have altered their shape, as in Fig. 230. Taking the centre plank first, it will be observed that the thickness at the middle remains unaltered, at the edge it is reduced, and both sides are rounded, while the width remains unchanged.
The planks on each side of this are rounded on the heart side, hollow on the other, retain their middle thickness, but are reduced in width in proportion to their distance from the centre of the tree; or, in other words, the more nearly the annual rings are parallel to the sides of the planks the greater will be the reduction in width. The most striking result of the shrinkage is shown in Figs. 231-233. Fig. 231 shows a piece of quartering freshly cut from unseasoned timber; in Fig. 232 the part coloured black shows the portion lost by shrinkage, and Fig. 233 shows the final result. These remarks apply more especially to oak, beech, and the stronger home firs. In the softer woods the medullary rays are more yielding, and this slightly modifies the result; but the same principles must be borne in mind if we wish to avoid the evils of shrinking which may occur from negligence in this respect.
The peculiar direction which "shakes" or natural fractures sometimes take is due to the unequal adhesion of the woody fibres, the weakest part yielding first. In a" cup-shake," which is the separation of a portion of 2 annual rings, the medullary rays are deficient in cohesion. The fault sometimes occurs in Dantzic fir, and has been attributed to the action of lightning and of severe frosts. So far we have considered the shrinking only as regards the cross section of various pieces. Turning now to the effect produced when we look at the timber in the other direction, Fig. 234 represents a piece of timber with the end cut off square; as this shrinks, the end remains square, the width alone being affected. If, however, the end be bevelled as in Fig. 235, we shall find that in shrinking it assumes a more acute angle, and this should be remembered in framing roofs, arranging the joints for struts, etc., especially by the carpenters who have to do the actual work of fitting the parts. If the angle be an internal one or bird's-mouth, it will in the same way become more acute in seasoning. The transverse shrinkage is here considered to the exclusion of any slight longitudinal alteration which might occur, and which would never be sufficient to affect the angle of the bevel.
When seasoned timber is used in positions subject to darnp, the wood will swell in exactly the reverse direction to the shrinkage, and induce similar difficulties unless this point has also received due attention. Of course it will be seen from a study of the cross sections illustrated in the diagrams that the pieces might be selected in such a way that the shrinkage and expansion would take place chiefly in the thickness instead of the width, and thus leave the bevel unaltered. In this consists the chief art of selecting pieces for framing; but in many instances motives of economy unfortunately favour the use of pieces on stock, without reference to their suitability for the purpose required.
It has been proved that beams having the annual rings parallel with their depth are stronger than those having them parallel with their breadth. Thus in the log shown in Fig. 236, the beam cut from A will be stronger than that from B. In preparing floorboards, care should be taken that the heart does not appear on the surface of the finished board, or it will soon become loose and kick up, as in Fig. 237, forming a rough and unpleasant floor. When planks which have curved in shrinking are needed to form a flat surface, they are sometimes sawn down the middle, and the pieces are alternately reversed and glued together, as in Fig. 238, each piece tending to check the curvature of the others.
In converting fir timber in Sweden and Norway, each log is inspected before sawing, to see how many of the most marketable sizes it will cut, and then it is marked out accordingly. The most general arrangement is that shown in Fig. 239, the thicker deals being for the English and the thinner for the French market. Another plan, shown in Fig. 240, has the disadvantage that the central deal embraces all the pith, and is thus rendered more liable to dry-rot.
In converting oak, the log is first cut into 4 quarters, each of which may then be dealt with as shown in Fig. 241. The best method is represented at a; it gives no waste, as the triangular portions form feather-edged laths for tiling, etc.; it also shows the silver grain of the wood to the best advantage, b is the next in order of merit; c is inferior; d is most economical for thick stuff.