Experiments in Composite Bridges – Part 1
(This is Part 1 of a two-part article. Conclusion next week.)
It is a little known fact that the heaviest load that most bridges carry is not the traffic that runs over them. It is the bridge’s own weight, the dead load. The bridge superstructure and deck often far outweigh the live load. But what if it didn’t have to be that way?
In the late 1990’s and early 2000’s, there was a good deal of experimentation being conducted by various state departments of transportation (DOTs) around composite bridges and bridge elements. At least two of these projects were complete bridges with FRP superstructure and deck, one in New York state and one in Iowa. Both were made using an interesting cellular design, built by Hardcore Composites LLC of New Castle, Delaware. (We’ll get to the technology next week.)
The bridge in New York replaced a crumbling concrete slab bridge on Route 248 just south of Rexville, NY. It’s in a rural area with a largely agricultural economy, and the bridge saw about 300 vehicles per day, 17% of them trucks including logging trucks and milk trucks. The old bridge, built in 1926, had been downgraded to a 10-ton limit in 1997 because of its deteriorated condition. This limitation had placed some strain on the community, forcing heavy trucks to go elsewhere. The original thickness of concrete had been increased by several asphalt repavings, and the deck depth was 900mm (3 feet).
The new, FRP bridge was considerably slimmer, and raised the clearance for water underneath by about 10 inches. It was made in two sections, each 7.8 m long and 5 m (one lane) wide. The original bridge was removed in March of 1998, the abutments rebuilt, and the new bridge placed in August, 1998. It took about six hours to install it.
The composite was made of e-glass fabric with vinyl ester resin. The bridge had a 10mm integral wearing surface made of polymer concrete.
Once installed, the bridge was instrumented with strain gauges, and then tested with four NY state-owned dump trucks. Its measured deflection was 3.5mm, substantially less than the design limit of 8.8mm. After being successfully tested, it was opened to traffic.
A Dec. 2000 report on the project by the Transportation Research And Development Bureau of the New York State Department Of Transportation concluded:
Fiber-reinforced polymer (FRP) materials can be a viable alternative for replacing short-span
concrete slab bridges in a cost- and time-effective manner. Since most components are shopfabricated, the time required for construction can be reduced significantly for small bridges. Even though FRP component costs are higher than traditional materials on a square-foot basis, they may be competitive in terms of life-cycle costs (including construction costs, user costs, maintenance costs, etc.).
The total cost of fabricating and installing the bridge was $400,000. It was estimated that a raditional bridge replacement would have cost $1,459,000.
I have not been able to find a report of the long-term fate of that bridge, but I have pieced together bits of the story.
- Records of the bridge inspection reports from 1998-2010 found it in “not deficient” condition. (That’s what it should be.) (This data is on a website called – no kidding – uglybridges.com)
- In 2009, NYSDOT announced that it would repair the deck.
- In 2010, both deck and superstructure were rated Satisfactory with a sufficiency rating of 97.7.
- A presentation dated 2011, by Jerome S. O’Connor, PE, F-ASCE, MCEER Sr. Program Officer, Transportation Research, University at Buffalo, reported problems with debonding of the polymer concrete wearing surface, showed 2009 pictures of a sample of the debonded top skin with insufficient wet-out of the fibers (a clear fabrication error) and an attempted repair of the surface.
It was reportedly reconstructed in 2012, and it is now listed as a pre-stressed concrete bridge. In 2012 and 2014 inspections, the new bridge had the same sufficiency rating, 97.7, as the FRP bridge had at the end of its service.
~ ~ ~
The Iowa bridge was a temporary bridge, portable, designed to be popped in place on a detour when an existing bridge is being repaired or replaced. Iowa had a number of similar structures made of steel, but they were getting old and in need of replacement, so they decided to experiment with one made of FRP.
The Iowa temp bridge was similar to the New York state bridge in technology, but longer in span and thicker in superstructure depth. It was built and first tested in 2005. It was built in two panels, each 39’10” long and 13’6.5’ wide. Each panel provided one 12-ft traffic lane and a railing. In service, they were spliced together with 1” thick steel plates top and bottom. The span of the bridge 39’0”, leaves just 5 inches at each end for bearing on the abutments.
The initial tests had the bridge set on temporary abutments in a DOT yard. An unusual preliminary test was conducted, loading it from underneath by placing a hydraulic jack under it and pressing upwards with a force equal to one half of a rear axle load of 7000 lbs. The deck was instrumented with strain gauges, and tested at multiple locations. It didn’t break, and deflected only minimally.
After that, they backed a loaded dump truck onto it. It deflected 0.34 inches, less than the limit of 0.59 inches.
It was installed on an actual jobsite in northern Iowa in the spring of 2007, with instrumentation for measurement purposes. However, shortly after traffic was allowed onto it, inspectors observed damage to the leading edge of one of the panels. The bridge had been installed 1 inch higher than the edge of the abutment, and the edge of the FRP top skin was getting hit by the full force of oncoming traffic. Steel plates were placed over the transition to protect the edge.
Shortly after that, they noticed that the top FRP layer and the wear surface had debonded over about half the surface area of that panel. (It is unclear from the report whether or not the debonded section was contiguous with the damaged edge.) In addition, water had gotten in through holes in the FRP around the lifting lugs used to install it, and it was feared that the foam cores might be deteriorating. The bridge was tested with a similarly-loaded dump truck on it, weighing 52,000 lbs. It deflected less than it had in the initial tests on the temp abutments. However, because of the debonding, after the tests, the bridge was immediately closed and removed.
The circumstances of the failure suggest that the debonding might have been caused, or at least hastened, by the lateral force of traffic hitting the exposed leading edge. If so, it seems like an unfair test, but then, that’s what the real world is like. (For more details, see FRP Bypass Bridge Final Report 2 by the Bridge Engineering center, Iowa State Univ.)