When a short piece of wood fibre, such as that shown in Fig. 6, is dried, it shrinks, its walls grow thinner
Fig. 7. - Warping of Wood.
(as indicated by the dotted lines), its width, a b, becomes smaller and the cavity larger, but the height or length, b c, remains the same.
The thinner the walls of the fibre the less also is the shrinkage.
The end walls of the fibres also shrink in the same way as the sides, but as the length is often a hundred or more times as great as the diameter, the effect of the longitudinal shrinkage is inappreciable.
A thin cross section of several fibres shrinks in the same way, the walls of each fibre become thinner and the whole piece contracts in proportion. Where the cells are very similar in size and thickness the piece shrinks by about the same amount on all sides, but if the piece is made up of fibres, some of which have thin and others thick walls, then the row of thick-walled cells, shrinking much more than the row of thin-walled cells, the piece becomes unevenly shrunk or warped, as shown in Fig. 7. Not only is the wood warped, but the force which led to this warping continues to strain the interior parts of the piece in different directions.
" Since in all our woods cells with thick walls and cells with thin walls are more or less intermixed, and especially as the spring wood and summer wood nearly always differ from, each other in this respect, strains and tendencies to warp are always active when wood dries out."
The pith or medullary rays also have a marked effect upon the shrinkage of wood.
As was shown in Section 2, the cells of the pith rays have their length at right angles to the direction of the wood fibres; hence as the pith rays dry they pull on the longitudinal fibres and try to shorten them, and, being resisted by the rigidity of the fibres, the pith ray is greatly strained. The fibres also shrink at right angles to the pith rays, and the latter in opposing this prevent the former from shrinking as much as they otherwise would. Thus the structure is subjected to two severe strains at right angles to each other, and it is principally owing to these strains that whenever the wood dries rapidly the pith rays separate and checks result, which, whether visible or not, are detrimental in the use of the wood.
"The contraction of the pith rays parallel to the length of the board is probably one of the causes of the small amount of longitudinal shrinkage which has been observed in boards." *
Owing to the opposing of the shrinking of the fibres in a radial direction by the pith rays, all woods shrink more in a tangential direction, or around the rings, than in a radial direction, and this greater tangential shrinkage affects every phase of woodworking.
The effect of seasoning upon a log is shown at A, Fig. 8. The external portions of the wood shrink the most and the heartwood but little, and the wood splits in radial lines from the centre parallel with the medullary rays, but maintains its original diameter.
* This longitudinal shrinkage, however, is so very slight that in practice it is customary to assume that the length of a timber is not affected by shrinkage.
Sawed in half, the log shrinks as shown at B, and if converted into boards by sawing in the usual way, the boards take the forms shown at D, all owing to the greater tangential shrinkage of the wood.
If the log is cut into four square timbers, one edge being in the centre, the pieces will shrink to the shape shown in Fig. 9. It will be seen that in . this case the diagonal, d c, remains unchanged, but the thickness of the timber each way is less and the angles are no longer square. Timbers sawn in this way, however, are much less liable to check than when sawn as shown at C, Fig. 8.
Timbers sawn as in Fig. 9 are known as quartered wood. A round shaft turned from a quartered timber before the latter had seasoned would shrink as shown in Fig. 10.
There is also a great difference in the effects of shrinkage in different woods. The soft woods, such as pine, spruce, cypress, redwood, etc., with their very regular structure, dry and shrink evenly and suffer much less in seasoning than the hard woods. Among the latter oak is the most difficult to dry without injury.
Small-sized split ware and quarter-sawed boards season better than ordinary boards and planks.
Fig. 8. - Effects of Shrinkage.
To avoid warping and checking all high-grade stock is carefully seasoned before manufacture.
When boards or planks which have shrunk to a curved form have to be used to form a flat board, they should be sawn lengthways into strips and glued together, the alternate strips being reversed as in Fig. 11. In this way the curvature in each piece becomes very slight, and the reversal of the alternate pieces causes each piece to be a check upon the shrinkage of its neighbors.
In constructing fine cabinet work large and thick pieces of wood should be avoided, and if required should be made up of a number of thin pieces glued together; large surfaces should be made in panels, or of smaller pieces covered with veneer.
Large timbers, when used for posts, are less liable to check if a 1-inch or 1½-inch hole is bored longitudinally through the timber. This hole should be connected with the outer air by ½-inch holes near the top and bottom of the post.
Large wooden beams and girders may be largely kept from twisting and checking by cutting the beam in halves through the heart of the log and bolting the two pieces together with the heart sides outward, as shown in Fig. 12.
Although only the first shrinkage is apt. to check the wood, repeated swelling due to changes of moisture increases the injuries produced in the first seasoning, so that wood should always be protected from moisture when once it is dry.
Bent Wood. - Steaming wood permits it to be bent more easily, and if the wood is bent before seasoning and kept in position until seasoned, it retains its bent shape and firmly opposes any attempt at subsequent straightening. 16. Amount of Shrinkage. - The shrinkage of wood varies for different species and even in different parts of the same tree; hence any figures that may be given for the shrinkage of wood must be regarded as mere approximations. Sap-wood, as a rule, shrinks more than heartwood of the same weight, but very heavy heartwood may shrink more than lighter sapwood. Quarter-sawed boards shrink less in width than those that are bastard sawed, but more in thickness.
The following table, given by Mr. Roth in the Government bulletin before referred to, is probably as reliable as any data that can be given:
VARIETY OF WOOD.
All light conifers (soft pine, spruce, cedar, cypress).............
Heavy confires (hard pine, tamarack), honey locust, box cider......
Basswood, birch, chestnut, horse chestnut, blue beech, young locust..
Hickory, young oak, especially real oak..........................
Up to 10