Built members latticed on four sides occur frequently in pure tension, contrary to the first principles of economy. A further saving might have been effected by the introduction of intermediate horizontal struts in the plane of the trusses, with a view to diminishing the effective length of the compression members and thus increasing the allowable unit stresses. It is believed that a gross saving of at least 20 per cent in the weight of the trusses might be made by modifying the design with an eye singly to economy. It is needless to state that such changes are not recommended. They could, in fact, not be made without altering the whole character of the design. METHOD OF ERECTION. The bridge may be erected in part by the cantalever method after constructing either the main arch or the half arches on false works. Since, however, the trusses are all riveted, it may be found preferable to use false works throughout. Whatever method should be used, no additional material will have to be provided for erection stresses. In the above estimate liberal allowance has been made for the weight of details in keeping with a strictly first-class structure. RAILING, CORNICE, AND ORNAMENTAL CASTINGS. Attention has already been called to the importance of substantial and ornate railings and cornices. These are essential to the good appearance of every highway bridge. In the case of a deck bridge they become in a sense the crowning features of the design. The cost estimated for these items (Appendix A) is believed to be sufficiently liberal, but not unduly extravagant. The general artistic effect may be materially heightened by the use of well-designed ornamental castings at the joints of the upper chord. These should be made of the same depth as the chord and about 4 feet in length, so as to completely cover the tie plates (Sheet I) between the latticed portions of the chord. PAVEMENT. The roadway pavement has a total thickness of 8 inches, consisting of a 5-inch cement concrete base, a 14-inch binder, and a 14-inch asphalt surface. The sidewalks have an asphalt wearing surface 14 inches thick, on a 31-inch base of bituminous concrete. Six-inch vertical offsets are provided between the levels of the sidewalks and roadway. GRADES AND DRAINAGE. For the proper drainage of the roadway, a longitudinal grade of 4 inches per 100 feet is to be provided in both directions from the center of the bridge, with the crown of the arch in its lowest position from temperature. This is to be accomplished by the insertion of raising pieces under the roadway joists (not under the transverse floor beams). Transversely the roadway pavement is to be crowned to a height of 73 inches at the center, by suitable variations in the heights of the raising pieces referred to. The sidewalk pavements are to be elevated 1 inch along their outer edges, so as to drain toward the roadway. Scuppers, with drainpipe connections to the sewers or to the creek, are to be provided at the ends of the bridge and over the extremities of the main arch. MASONRY. End abutments.-The end abutments are designed as U-shaped, with concrete foundations below the ground line, extending to solid rock, as shown on Sheet III. The masonry above the concrete is to consist of granite-coursed ashlar, quarry faced, with chisel-draft edges. Arch abutments.-The proposed arch abutments are shown in elevation on Sheet IV. The east abutment is designed of grauite ashlar throughout, the west abutment of granite ashlar on a concrete base. The masonry below the horizontal coping is to be finished as described above for the end abutments. In determining the pressure lines of greatest intensity and obliquity, the following conditions of live loading were considered (H denoting the horizontal and V the vertical component of the reaction): (a) Live load in position for maximum positive V. Combining these values with the dead load reactions, the five resultant pressures are shown in magnitude and in position on Sheet IV, the notation indicated corresponding to that used above. The masonry is to be laid on inclined beds, approximately normal to the direction of the line of pressure (e), for the bridge wholly loadedi. e., parallel to the upper surface of the concrete base, as shown on Sheet IV. The outer courses of the stonework may be bedded hori zontally, for the sake of better appearance, as represented in the general elevation of the bridge on Sheet I. In Cooper's Specifications, the allowable pressure per square inch on the bedplates is fixed at 250 pounds. It is recommended that this be increased in the present case to 300 pounds per square inch, or approximately 22 tons per square foot, first, because of the unusually high ratio of the dead to the live load, and, secondly, because it may be safely assumed that the computed live reactions are considerably greater thar will be realized in the lifetime of the structure. The computed maximum pressures at the joints in the masonry are about 15 and 7.5 tons per square foot on the granite and concrete, respectively, both values being well within the limits of safe practice. The estimate of cost is based on natural cement concrete throughout. Estimate of quantities.-The estimate of the required volume of masonry should be regarded as only loosely approximate. Shafts for determining the depths to hard rock were sunk only along the center line of the proposed site. The lateral slope of the rock can only be conjectured. The following estimate of quantities is based on the center profile, with depths as shown on Sheets III and IV: A rough estimate of the necessary excavation for the masonry, on the basis of the center profile, is given below. No account is taken of the grading under the bridge, nor is this item included in the estimate of cost appended to this report. To prevent possible injury to the steel work during floods, the masonry must be carried somewhat above the extreme high water level. The following figures will show that adequate provision has been made against danger from this cause: The end-pin centers of the arch are at an elevation of 34.1 feet above datum. The area of clear waterway between the proposed profile (Sheet I) and the 30-foot level is approximately 2,875 square feet. The highest measured freshet discharge at the mouth of Rock Creek, according to the report of the Engineer Commissioner, District of Columbia, under date of January 10, 1893, was at the rate of about 10,000 cubic feet per second. It is stated in that report that "provision should be made for about double that rate." A discharge of 20,000 cubic feet per second for a cross section of 2,875 square feet corresponds to a mean velocity of about 7 feet per second. This agrees substantially with the mean velocity computed by Kutter's formula, assuming a slope of 1 in 1,000 and a coefficient of roughness of 0.03. COST ESTIMATE. The cost estimate for the structure, complete, as shown in detail in Appendix A, is $199,204. This estimate has been made on a safe basis throughout, and will insure a strictly first class structure. It is believed that the actual cost of the bridge, if constructed in the near future, will be somewhat less, rather than greater. Thus the contract for the steel work could doubtless be placed somewhat more advantageously than estimated, under the abnormally low prices prevailing at the present time. MODIFICATIONS FOR ECONOMY. A reduced cost estimate is presented in Appendix B. This reduction is effected by the omission of the cornice and the ornamental castings along the upper chord, and by the adoption of a cheaper railing. These modifications are not only not recommended, but should, in my judg ment, be condemned on the ground of false economy. The omission or partial suppression of these important ornamental features would inevi tably result in imparting to the whole structure an appearance of cheapness, with an attendant saving in cost of only 8 per cent. An additional saving of about $2,500 might be made by reducing the thickness of the buckled-plate flooring from three-eighths inch to fivesixteenths inch. I would strongly advise against such a change. These plates will necessarily be weakened in time by rusting. Their upper surface does not admit of inspection or repainting, and their renewal is a very expensive operation. A considerable reduction of cost might further be effected by the substitution of wooden floor joists and planking for the steel beams and asphalt paving. The estimate, in Appendix B, of $183,304 may thus be safely reduced to $150,000. Such a change would require a complete modification of the general design. The reduction in the dead load would be, roughly, 6,600 pounds per linear foot of bridge for the floor system alone. This would necessitate an increase in the length ratio of the suspended trusses to the cantalever arms, or provision, by anchorages at the arch abutments, against negative horizontal reactions developed under special conditions of live loading. The adoption of an ordinary wooden floor could be warranted only on the basis of pressing economy. Such a construction would fail to meet the reasonable requirements of the surroundings, and would be entirely out of harmony with the design itself. ACKNOWLEDGMENTS. In conclusion, permit me to make my appreciative acknowledgments of your unvarying interest in the development of this design, and your promptness in supplying all data required. My thanks are due also to Mr. A. Y. Sundstrom and Mr. William Kleefeld, jr., students in the department of civil engineering at the University of Pennsylvania, for their efficient services in the preparation of the drawings accompanying this report. Very respectfully, Capt. D. D. GAILLARD, EDGAR MARBURG, Consulting Engineer. Corps of Engineers, U. S. A. Engineering, office, inspection, superintendence, and contingencies.. Estimated total cost... 200 80 200 35, 435 15,000 199, 204 I most heartily concur with Professor Marburg in the opinion that modifications of the design, with the view of decreasing the cost, are not to be recommended, as all such modifications detract from the appearance of the bridge. Should the steel bridge be constructed it is most important that the character and harmony of the design should be preserved throughout, and to accomplish this result I would most respectfully recommend that the designer be employed as consulting engineer whenever advisable during the construction of this bridge. CONCLUSIONS. The avenue in question is one of the most important in the city, and is terminated very abruptly on the northwest by the deep ravine in which Rock Creek flows. No city bridges span this ravine within half a mile of the site of the proposed bridge; consequently a bridge at this locality is much needed. Such a bridge should be of a character in keeping with the importance of the avenue on which it is situated. |