Return of the Wooden Bridge

Anne T. Smith
March 01, 1996

Wooden bridges and computer models seem to have as much in common as gas lights and CD players. But not to Penn State graduate student Ray Witmer.

For the Pennsylvania Department of Transportation, Witmer is developing a computer model of a modern wooden bridge under study as a possible replacement for aging concrete and steel spans. He gestures toward tall stacks of lumber lining one wall of his lab. Each layer is a single 10-by-4-foot plank about 6 inches thick, created by gluing together many smaller boards. "These are bridge parts," he says enthusiastically. "I'm going to bend them to generate data."

A prototype bridge was constructed in 1991 in Baileyville, five miles west of State College, PA. The only feature that distinguishes it to the casual traveler is the heavy wooden guardrail on either side of its asphalt deck. To see any of the wooden substructure requires stopping along the roadway, climbing down into the small stream below, and looking up underneath the span.

Such bridges "can carry maximum highway loads," says agricultural engineering professor Harvey Manbeck, Witmer's adviser, "and last as long as their steel or concrete counterparts." A photo on the wall in Manbeck's office illustrates their load-bearing capacity, showing two 40-ton dump trucks parked side by side in the center of the prototype bridge.

Manbeck has overseen the Timber Bridge project since 1989. PennDoT's interest in an alternative to concrete and steel was piqued, he recalls, by an American Society of Civil Engineers program to expand the role of wood as a structural material. Yet, while most timber bridge projects throughout the country relied on softwoods like pine and fir, with Pennsylvania the Northeast's top producer of hardwoods, Manbeck and PennDoT decided to develop a hardwood bridge. The result, says Manbeck, "is a renewable, local resource, competitive pricewise with concrete and steel."

Witmer's bridge is made using non-furniture grade red oak, red maple, or yellow poplar ("weed trees" because of their small size and abundance). The boards are kiln dried, cut to uniform widths, planed to an inch thick, then glued to create up to 90-foot stringers from boards only 8- to 12-feet long. Laminating distributes strength-reducing defects, such as knots and sloping grain, throughout the inner layers. Using higher quality wood in the outer laminations brings the strength of the finished timber up to that of the strongest boards.

The decking is glued into 4-foot-sections as long as the width of a two-lane bridge. Each section is laid across the stringers, bolted down, and joined to the next section by steel dowels or bars fitted into pre-drilled holes. A heavy polymer sheet is laid on top, then paved over with blacktop or concrete. The integrity of the paved layer—and the life of the bridge—depends on predicting how the bridge's wooden substructure will move under loads. Too much flex and the pavement-membrane layer will crack, letting water seep into the wood beneath.

Witmer's computer model will quantify the deflections where the stringers and decking meet, as well as in the decking-to-decking connections. It will allow designers to look at such variables as span length, stringer placement, and the angle at which the bridge meets the roadway (called the skew angle), all of which can be varied.

Currently, using a commercially available program, Witmer has created a mathematical representation of the prototype bridge. But to be sure his equations are correct, he needs actual data from real bridge parts under real loads. That's where the stacks of lumber in the wood engineering lab come in. While loading and measuring deflections in his laminated bridge parts, Witmer will also verify the tests themselves using data taken at the same time as the photo in Manbeck's office. When those dump trucks were parked on the prototype span, deflection measurements were taken at 1- to 2-foot intervals along the length of each stringer. These numbers, compared to Witmer's data, will tell the researchers if their model is accurate.

Ray Witmer is a Ph.D. student in the wood engineering program, an interdisciplinary degree program in the College of Agricultural Sciences. His adviser, Harvey Manbeck, is professor in the department of agricultural and biological engineering, University Park, PA 16802, 814-865-4071. Other members of the interdisciplinary Timber Bridge research team are John Janowiak, Paul Blankenhorn, and Peter Labosky Jr., all faculty in the School of Forest Resources. Ann T. Smith is a freelance writer..

Last Updated March 01, 1996