37 



The writer agrees with the author of the paper that the bearing 

 surfaces at the ends of stringers and of caps in railroad trestles are 

 usually too small, and that in consequence indentations of the timber 

 take place, which means a destroying of the fiber and, in connection 

 with moisture and constant working, subsequent speedy decay. But it 

 will be impossible to follow the author of the paper to the extent of 

 recommending the introduction of corbels under the ends of stringers. 

 This may be due in part to the individual objections that the writer 

 has always had against the use of corbels. These objections have not 

 been overcome by a perusal of the points presented by the author of 

 the paper in favor of corbels, and an examination of the design recom- 

 mended, showing corbels, fails to reveal wherein the bearing surface 

 on top of the cap is improved by the introduction of a corbel under 

 each stringer, considering that the corbels as proposed are of the exact 

 width as the stringers. The author of the paper must either assume 

 that the corbels reduce the clear span or else that only the panel on 

 one side of the bent is loaded. Both assumptions are wrong from a 

 practical standpoint. As above stated, it is impossible for the writer 

 to see how a 7f-incb corbel resting on a 9-inch cap. giving about 70 

 square inches of bearing surface on top of the cap, can be an improve- 

 ment over a stringer of the same width as the corbel, resting on a 

 12-inch cap, giving 93 square inches of bearing surface, or, according 

 to the " example of present practice," over an 8i-inch stringer resting 

 on a 12-inch cap, giving 99 square inches of section. 



The introduction of corbels is objectionable on account of the extra 

 cost, without any commensurate improvement of the design, and owing 

 to the undesirable feature of bringing more timber against timber, 

 especially on horizontal and boxed surfaces, increasing the liability of 

 decay. The recommendation (?) of the author of the paper to use short 

 stringers in place of double-length stringers is a disadvantage; while a 

 screw bolt through the corbel and stringer, tending to increase the 

 resistance of the beam to failure by shearing, would prove, if introduced 

 solely for this purpose, a foolish precaution in place of a wise one, for 

 the reason that with the seasoning of the timber the grip of the bolt 

 would be lost unless kept regularly tightened up. To do this to such 

 an extent as to be of practical benefit would be impracticable. There 

 are cases where corbels laid flat, i. e.. made wider than the stringers, 

 will increase the bearing value on the cap. There are also cases where 

 the elevation of the rails on curves on trestlework can be best accom- 

 plished by the introduction of corbels, giving an opportunity to vary 

 the elevation according to the thickness of the corbel. But as a steady, 



This latter arrangement allows the braces to be attached at their intersection by means 

 of a bolt. Hence, instead of computing a strut large enough to transmit the thrust 

 along the full length of the diagonal, about 18 feet, we need only determine the area 

 necessary to transmit it half that distance. That is, in the first case we have a 

 column 18 feet long, and in the latter case columns only 9 feet long, so that they do 

 not need to be so large. — A. L. Johnson. 



