The details of the simple wooden girder have already been considered, but it sometimes happens that it is necessary to use a girder of longer span than would be safe for the deepest single beam that can be obtained, and in such case, where steel beams cannot be obtained without great expense, compound wooden beams may be used.
By a compound wooden beam or girder, is meant a beam built up by placing two or more single beams, one on top of the other, with the view of having them act as a single beam having the depth of the combined beams.
This, if two 10-inch x 10-inch beams were placed one on top of the, other, and the upper one loaded at the centre, the beams would act two separate beams (Fig. 486) and their combined strength would be no greater than if the two beams were placed side by side. If, however, the two beams can be joined so that the fibres of the lower beam will be extended as much as would be the case in a single beam of the same depth, or, in other words, so that the two beams will not slip on each other, the compound beam will have four times the strength of the single beam.
Various attempts have been made to join beams thus placed, so as to prevent the two parts slipping on each other, but until very recently there has been no experimental data to show how far such methods accomplish their object. During the years 1896-7, however, Prof. Edgar Kid well, of the Michigan College of Mines, made quite an extended series of tests of the efficiency of compound beams of different patterns, and from these tests much valuable data has been obtained. A full description of the tests accompanied by the conclusions of the author, and rules and data for proportioning the bolts and keys, of keyed beams, is published in Vol. XXVII., Transactions of the American Institute of Mining Engineers.
Probably the most common form of compound beam, as used in
American building construction, is that shown in Fig. 487, diagonal boards in opposite directions, being nailed to each side of the two timbers to prevent their slipping on each other. Mr. T. M. Clark, in his Building Superintendence, advocates this as one of the best forms of compound beams, and places its efficiency at about 95 per cent, of a solid beam of the same depth.
Prof. Kidwell made nine tests of this style of beam, six having a ratio of span to depth of beam as 12 to 1, and three as 24 to 1. The shorter beams gave an average efficiency without much variation, of _ 71.4 per cent., and the longer beams an efficiency of 80.7 per cent.
It was found that the beams failed by the splitting of the diagonal pieces or the drawing of the nails - "in every case, long before the beam broke, the struts split open or the nails were drawn partly out, or bent over in the wood, thereby permitting the component beams to slide on each other. It was found that no amount of nailing could prevent this."
When built with diagonal boards 1 ¼ inches thick, nailed with 10d's, as in Fig. 487. the working strength of such a beam may be taken at 65 per cent. of the strength of a solid beam of the same depth, 2nd of a breadth equal to the breadth of the timbers. The deflection of the beam, however, will be about double that of a solid beam of the same size, and on that account this style of beam is not to be recommended for supporting floors with plastered ceilings or carrying plastered partitions.