California's Temporary Freeway Bridge
On the evening of March 10, 1995, in soggy Southern California near Coalinga, the twin bridges carrying Interstate 5 over the Arroyo Pasajero collapsed, killing seven people. The sandy-bottomed arroyo, which is normally dry, was no match for the swiftly flowing floodwaters, and the resulting scour around the bridge foundations led to their failure.
The California Department of Transportation (Caltrans) was faced with the dilemma of another closed freeway. Teams of bridge engineers responded immediately and worked over the weekend to develop strategies for opening the freeway. In a state plagued by disaster after disaster, the scenario of long hours and emergency contracts was almost commonplace.
Fortunately, the solution to this predicament was only 80 kilometers away at a Caltrans maintenance facility, where a new type of temporary bridge - made from surplus railroad flatcars - was stored. The idea of using railroad flatcars for bridges is not new. There are 83 such bridges in California. Some of which date back to the 1930s. These bridges are typically located on low-volume, local roadways.
However, the concept of using railroad flatcars as modular, reusable bridges - the brainchild of W.H. Wattenburg of the Lawrence Livermore National Laboratory - first came to Caltrans' attention following the 1994 Northridge earthquake. Although a railroad flatcar was not used to restore essential freeway service in the aftermath of the Northridge quake, Caltrans decided to pursue the idea for future needs.
Caltrans hired a bridge contractor to build a prototype flatcar bridge based upon Wattenburg's concept. The prototype was constructed entirely of flatcars, as shown in figure 1. A flatcar acts as a foundation and supports the half-flatcars that serve as columns, which in turn support a flatcar that acts as a bent cap. The deck system consists of four flatcars, interlocked side by side. The entire assembly is modular and versatile. In addition, it is structurally strong and passes the dynamic analysis performed by Wattenburg. (1)
Following the testing of the prototype, Caltrans stored sufficient modules to build a temporary bridge that could span a 50-meter (m) gap. The failures at Arroyo Pasajero provided an opportunity to use the modular bridge technology. For this site, Caltrans needed a three-span bridge with a deck that was four rail cars wide. The 12-car deck was 12.2 m wide and 48.9 m long, and it allowed one lane of traffic in each direction.
However, as it is often the case, the real-world application was not as straightforward as its prototype. First of all, it was quickly realized that the flatcar support system would not be practical in the river channel. The new substructure design consisted of steel H-piles, six per bent, braced with angles with 152-mm-long, 19-mm-thick legs; the flatcars were used only for the superstructure.
The second problem encountered was the variation in section thickness between the cars themselves, which necessitated the use of steel shims to acquire a uniform roadway profile grade.
Caltrans chose to use an open steel-grating system to furnish a roadway surface. Attaching the grating was another obstacle because eight of the 12 flatcars had thick, track plates welded to their tops, causing the surface to be uneven. Holes were cut in the grating to place it around the protruding plates, and surface plates were welded over these openings to compensate for the resulting instability in the grating. This adjustment added two days to the job.
The bridge was opened to traffic at 5:10 a.m. on March 18, 1995, eight days after the mainline bridges collapsed. Slightly more than 20 hours later, the deck grating system, which was inadequate for the heavy freeway traffic loading, began to fail. Caltrans closed the bridge and placed 25-mm-thick trench plates over the entire surface. Installing the 80 plates, each of which is 2.4-m by 1.2-m, took 12 hours. Penetron, a rubberized asphalt, was placed over the plates with a coating of sand for roughness.
The bridge was reopened to traffic at 12:30 p.m. on March 20, 1995, and remained open for another 29 days with no further problems. Much of the cost and delay associated with using the modular flatcar bridge technology at the Arroyo Pasajero was due to unanticipated problems for which field solutions had to be devised. Some of these costs could have been reduced and time saved if the cars had been modified in advance by welding shims for uniform depth, and by attaching the deck grating system. The following costs should be anticipated for this type of bridge:
- Hauling the flatcars to the site.
- Using cranes to place the flatcars. In an emergency situation, it may be necessary to use two cranes - one at each end of the bridge.
- Field welding, or making the pieces fit together.
- Dismantling and removing the bridge when it is no longer needed.
Before the bridge's trial in the field, Caltrans estimated the cost of a 50-m bridge at approximately $263,000. This could be reduced to $150,000 if a concrete deck was used in place of the steel grid (2). Now that the bridge has undergone its first field installation, the estimated costs are approximately $19,000 per flatcar with steel decking and modification costs included. For a 12-flatcar system, such as at Arroyo Pasajero, this amounts to $228,000.
However, the intangible benefits, such as the savings in user costs, should also be taken into account. Considering the benefit of having the temporary bridge in place to avoid the cost of detouring the freeway traffic during the construction of the permanent bridges, Caltrans estimates a net savings of about $500,000 at the Arroyo Pasajero site.
In general, using railroad flatcars for temporary bridges is a workable approach once the necessary modifications and refinements are made. The system is more than capable of carrying legal highway loading and is adaptable to virtually any length and width.
References
(1) W.H. Wattenburg, D.B. McCallen, and R.C. Murray. "A Modular Steel Freeway Bridge: Design Concept and Earthquake Resistance," Science, Vol. 268, April 14, 1995. (2) James Roberts. "A Temporary Bridge," Science, Vol. 268, April 14, 1995.
Nancy McMullin Bobb is the division bridge engineer in the California Division of the Federal Highway Administration. She received her bachelor's degree in civil engineering from the University of Nevada-Reno and her master's degree in civil engineering from the University of California-Davis.