While very few exist, the Paddleford trusses are remarkable in that the assembly of interconnected timbers requires exceptional skill for a proper fit. These structures behave more like frames than trusses, involving shoulder bearing at the frame connections with much of the resistance due to shear and bending stresses in the elements, in addition to the axial forces. The analysis of these structures is especially complex and challenging.
Peter Paddleford (1785-1859; active 1820-1850) was a millwright and bridge builder in northern New England who developed his non-patented Paddleford truss design after experience building Long and Pratt trusses. While little is known about his life and career, Joseph Conwill has observed that his truss "became the dominant type in covered bridge construction over a wide area extending from Orleans County, Vermont, eastward across northern New Hampshire, and on through Oxford County, Maine." The reason for this broad distribution is probably due more to local builders copying, or trying to copy, Paddleford's design, than to the work of Paddleford or his son Philip.
The Paddleford truss has the appearance of a multiple kingpost truss reinforced with distinctive counter braces. In trying to understand Paddleford's conception of his design's structural behavior and in trying to understand how it actually behaves, it is useful to consider another possibility. Paddleford could have conceived of his design as a non-prestressed variant of a Long or a Howe truss. The counter braces become eccentrically placed diagonals from this perspective.
In examining each of these possibilities, it is important to keep in mind the redundancy of the design, the nature of the connections, and how the top and bottom chords carry forces in the counter braces. A typical Paddleford truss is highly indeterminate. It has many more members and connections than needed for stability and consequently, there are many load paths. The configuration of the connections between the counter braces and the other truss members makes it clear that Paddleford envisioned a tension member. That is, if Paddleford expected compression in the counter braces, he would have used the type of butt connection found between the braces and the kingposts - what Dario Gasparini has termed "non-positive." Finally, it is important to realize that any vertical forces transmitted between the chords and the counter braces involve the bending strength of the chords. Applying the method of joints to the chord-counter brace panel points makes this clearer. Static equilibrium requires a mechanism to resist the vertical components of the forces in the counter braces. Since the connections lack another member to do this, bending in the horizontal chords must resist the vertical forces. Depending on the placement of the deck beams, the bottom chord might transfer a significant amount of bending moment to the counter braces. University of Vermont engineering professor Jean-Guy Beliveau, however, has suggested that the counter braces would not impose much bending moment on the top chords.
Given his documented experience building Long trusses, Paddleford may have sought to improve on certain deficiencies he observed -- an impulse not unlike the one that motivated Howe and Stone in their development of the Howe truss. Long's truss employs wedges to impose sufficient prestress on the diagonals so that they remain in compression for any live loadings. Certain load placements put tensile loads on some of the diagonals, but proper prestressing generates enough compression that the net force will still be compression. This condition permits the use of simpler, non-positive, connections for all diagonals. Paddleford's experience with Long trusses would make him familiar with these concepts, but he chose not to employ them on his trusses. (The reader will remember, however, that the upper lateral bracing uses wedges.) Surely Paddleford realized that a structural member placed in the general area of his counter braces would take tensile forces. The question might then be, what were his options in adding a tensile member to a multiple kingpost truss? Whatever he might have chosen, he could not reduce the critical section at the connection between the post and the chords because to do so would weaken a tensile member. If Paddleford conceived of his truss as a non-prestressed improvement on Long's work, then the eccentrically placed counter brace was one way to do it.
Another factor that Paddleford might have considered, whether he conceived of his design as a Long or a multiple kingpost, is the dimensional instability of wood. Timber framers require wood that is green enough to work easily. As wood dries and hardens, however, it shrinks. On the other hand, wood is also subject to creep under prolonged loading. Creep shortens compression members and elongates tensile members. Together, shrinkage and creep can loosen critical connections. Thus both Justin M. Spivey and John Ochsendorf raise the possibility that Paddleford might have seen his counter braces as a means to restrain the members and 99 compensate for these dimension changes. Unlike the Long and Howe trusses, Paddleford's arrangement would not require post-construction adjustments. Spivey observes: "Whereas a Long truss has (and needs) prestressing wedges to keep the compression counterbraces active, the Paddleford truss has tension counter braces that are always active, no matter how much creep deformation occurs, until failure occurs. The Paddleford truss form is more intriguing because the counterbraces and primary diagonals are connected to each other and restrain each other's creep deformations. Could the "primary" diagonals lose contact with the vertical post shoulders and the counter braces start carrying dead load? Or do the multiple connections between web members work to prevent this somehow? A surprisingly complex analysis would be required to obtain satisfactory answers to these questions."
The alternate view, that Paddleford conceived of his design as an improvement to the multiple kingpost truss, is much simpler and more direct. Paddleford may have been seeking to strengthen a successful design by adding additional load-carrying capacity. Perhaps he added the counter braces to reduce bending of the bottoms chords due to deck beam loads. In this case, any connection restraint was an added bonus. Tom Peters has observed that the overlaying of familiar truss types in order to create longer and stronger bridges was common in at least the eighteenth and nineteenth centuries. If so, then Paddleford is not working within the emerging academic tradition as much as within the craft tradition. Certainly, his connections require much more framing skill than the more scientifically based Howe truss.
The Paddleford truss never became as popular as any of the patented trusses. And, many examples of the truss have since disappeared, including all of the bridges attributed to Peter Paddleford himself. Today, there are only twenty-two remaining Paddleford truss bridges in the world, five in northwestern Maine, fourteen in northern New Hampshire, and three in northeastern Vermont.