Not every wooden trussed arch was a Burr truss
The first iron bridge was constructed in 1779 at Coalbrookdale, a cast-iron arch. Cast iron then began to replace stone, quite successfully in arch bridges, but the attempt to use cast-iron beams ended in the failure of Robert Stephenson's Dee Bridge. Thereafter, cast iron was used for compressive members only, and engineers turned to wrought iron as a tougher, more reliable material, using it first in massive tubular girders riveted together from the small plates which were all that were available at the time. Wrought-iron link chains were used for suspension bridges by Telford and Brunel.
In the United States, iron was expensive and largely imported from Britian because of the primitive state of the domestic iron industry. Wood, however, was abundant and cheap, and it was good hardwood that made an excellent material of construction. Wood, therefore, was the material most often used for even major bridges, with spans greater than 50 feet. We are not talking here of the trestles and king-post trusses used for minor bridges, but bridges that crossed the rivers and gorges that abounded. Two kinds of bridges were in common use, the lattice truss and the trussed arch, of which there were as many varieties as there were builders. Bridge builders were self-taught craftsmen and amateurs, for the most part.
Wooden bridges are exposed to the elements, and unless protected are quickly attacked by decay, especially at the critical connections between members. It was, therefore, usual to roof and sheathe the bridge to protect it, resulting in the "covered bridge" that was known to a later era by the surviving wooden bridges on back roads. However, this form of protection for wooden bridges was almost always applied, to large bridges as well as small. Fire and decay are enemies of all wooden structures, and one or the other generally ended the life of every wooden bridge. Wood has one outstanding advantage, that it complains audibly before it fails, giving a warning that something is amiss. Trautwine tells us that wooden and iron bridges of equal strength are of approximately equal weight. This means that wood has no advantage in dead load, because it is weaker than iron, weight for weight.
Railways loaded bridges more heavily than had been usual in the past. For the first time, bridges had to support a heavy, concentrated moving load. Although bridges were strengthened to assume the heavier loads, there was no theory and no rational design available, so matters quickly surpassed existing experience, and many of the bridges collapsed under load. This was especially true of the lattice trusses, which were propped up and braced in an attempt to strengthen them, and finally given up as a bad idea. Most wooden bridges of longer spans consisted of a wooden arch, stiffened by a truss consisting of upper and lower chords connected by posts and diagonal braces. Haupt presents examples of the best of these bridges, which were already beginning to use wrought iron for ties, and cast iron for connections, together with other improvements.
The association of a truss with an arch was not mere artistic choice. The arch is a very efficient structure, but there are significant differences between masonry and wood arches. A masonry arch is very heavy, so that the moving load is only a small part of the total load. The weight in the spandrels keeps the arch from distorting under a moving load, so that it does not fail by buckling. On the other hand, a wooden arch is light, and the moving load is the greater part of the load. The arch tends to fail by buckling when the moving load is applied only on part of the bridge. The part under the load falls, and the opposite end of the arch rises, distorting the arch. This is illustrated in the Figure. This tendency must be resisted, either by a stiffening truss or diagonal ties.
This action was not well understood by American bridge builders, who tended to think of the truss and arch as separate load-bearing structures that acted in concert, each component assuming its part of the load. The problem was that this sharing could not be guaranteed to be fair, because the different elements worked under different principles, and deflected differently under loads. Burr, who designed a trussed arch, and Wernwag, who designed a braced arch, seem to have understood the problem better than most, and their bridges proved quite successful.
It was impossible to make the arch and the truss work together to share the load equally, so unless each of the two elements was alone capable of supporting the load, the bridge was subject to collapse, and many did so. Haupt concluded that the arch should be capable of the entire load, and the truss used only to stiffen the arch for moving loads, a principle of the Burr design. This gives by far the most economical bridge. He correctly pointed out the nature of moving loads as creating reversed stresses, and emphasized the importance of counter bracing, which had usually been neglected up to that time.
The best of the trussed arches was the Burr truss, the most widely used bridge, at least in Pennsylvania, if not in the whole country, at that time. The name is a misnomer, since it is principally an arch, not a truss. Haupt gives as an example the bridge across the Susquehanna at Clark's Ferry. This is a dual-carriageway road bridge with an outside towing path on one side. The vertical posts transfer the load to the arch, and the diagonals are in compression, and there is no counter bracing. From the Bill of Timber, the bridge appears to have a span of 40' 6", and is roofed and sheathed with 22,260 shingles. The true, original, Burr truss is made completely of wood, including the trenailed connections. The arches spring from points below the lower chord of the stiffening truss, and are made of laminated timbers.
The Burr truss is also mentioned quite favorably in Trautwine, at a date when iron and steel bridges had nearly driven out wooden bridges, except for trestles. The diagram below (from Trautwine) shows the general pattern of the bridge, as well as the use of iron rods as ties, and the details of some of the connections. Trautwine repeats Haupt's dictum that the arch should be capable of supporting the whole load. He does not show counter bracing, but mentions that it is necessary for a railway bridge because of the heavy moving loads.
A wooden arch was often stiffened with a Long truss, which was a rectangular wooden truss with iron diagonals in each panel. The Long truss was invented by Stephen H. Long (1784-1864) of the U. S. Army Topographical Service, also a designer of useless locomotives, and a West Point graduate before its reformation. The Pratt truss, which looks quite similar if the diagonal ties are alike, was also used. If the vertical posts in the Burr truss, which are actually tension members, are replaced by iron rods, we have essentially the Howe truss, a very successful and strong design. The Howe truss received undeserved criticism when there were notable failures of this kind of bridge. The failures were due, however, to shabby construction, poor connections, and bad design, not to the basic form of the bridge. In competent hands, it served excellently. A wooden arch stiffened by a Howe truss was a very sturdy and safe bridge, that can be regarded as a descendant of the Burr truss. Haupt used such bridges on the Pennsylvania Railroad with complete success. An excellent illustration of such a brige is found in Alexander: the Allegheny River bridge at Freeport. There are five spans, each with four timber arches, each pair of arches sandwiching a Howe truss. If the date of 1853 for construction is correct, this bridge served for thirty years, an excellent record for a wooden bridge.
Haupt gives extensive details of the Pennsylvania Railroad bridge across the Susquehanna at Rockville, finished in 1849. Each span of the 23-span, 3670-foot-long bridge was a wooden Howe truss with iron ties combined with arches, and here either the truss or the arch was capable of supporting the whole load by itself. This bridge, too, was completely sheathed, at least most of the time. Later photographs showing it without sheathing were taken before the bridge was dismantled, and an early lithograph also shows it exposed. Each span was 160 ft center to center of piers, with 16 panels, and 18 ft high. Each panel had counter diagonals. The wood was mainly White Pine, which cost only $13 per 1000 board feet. The vertical iron ties were called "bolts" and ranged from 1-3/4" to 1-1/8" diameter, 18' 7" and 19' 6" long. Similar ties suspended the load on the arches. The arches rose 20' 10", and had a cross-sectional area of 1044 sq. in. at the middle of the span. This was a deck bridge, and the track was on top of the upper chord of the truss. The bridge was very substantial, and represents the best in wooden bridge construction for heavy loads. It is not, however a Burr truss, but something rather different. It was replaced by a double-track iron truss in 1877. This truss seems to have been a modified Pettit truss, with no posts and diagonals carried through from upper to lower chord. It was replaced in turn by the present four-track stone arch bridge in 1902.
Another famous bridge, built much earlier, is often erroneously referred to as a Burr truss. This is the Trenton Bridge across the Delaware, built by Lewis Wernwag in 1804. It was actually a very substantial wooden arch bridge with five arches, between which were two carriageways (11' each), and two footpaths (6' each). The arches were composed of eight 4 x 12 planks in contact, and the roadway was supported by chains of 1-1/8" square iron, with links 4' long and 5" wide. There was no truss, but the diagonal counter bracing may have given this impression. There were five spans. This bridge carried railway coaches in 1835, and was strengthened to carry locomotives as well in 1839. It did the job until it was finally replaced by a through iron Pettit truss in 1875. It was photographed in that year, when it had served for 71 years. Incidentally, the replacement bridge was iron, not steel, because steel was not yet trusted in truss bridges. Its design may have not been satisfactory, since a steel Pratt truss bridge soon came, in 1877 says one caption, and was in turn replaced by a stone arch in 1903.
American wooden bridges have received a bad reputation. Since ignorant design and careless workmanship were widespread in the United States, this reputation is probably not without justification. The use of the combined arch and truss has been viewed with particular contempt. However, this opinion is due to the same lack of understanding that was evidenced by the early builders. I hope I have demonstrated above that a wooden arch is an economical and strong structure, but requires to be stiffened to make a satisfactory bridge. Theodore Burr and Lewis Wernwag, and the competent engineers who followed them, such as Herman Haupt, designed sturdy and long-lasting bridges that should by no means be depreciated. By 1890, all of these bridges were gone, replaced in the most part by statically determinate steel Pratt trusses. In 1877, the Susquehanna bridge was the only remaining wooden bridge on the Pennsylvania main line. Another reminder of the early United States had disappeared.
Composed by J. B. Calvert
Created 12 October 2000
Last revised 23 October 2000