U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-04-098
Covered bridges are no different than other bridges when it comes to designing foundations for new structures or replacing abutments for existing bridges. This process requires soil borings, evaluation of bearing pressures, rock (if encountered), evaluation of the need for piles, and scour protection. This chapter focuses on the evaluation of existing foundations, and remedial action as required.
Most covered bridges are single-span structures with two abutments, one on each bank of the stream. Piers or bents, when present, are placed between abutments. Almost invariably, covered bridge abutments built in the 19th century were made of stone that was either mortared or laid dry (without mortar). Many of those foundations remain in service today. Figure 67 depicts a relatively tall stone abutment in good condition. In other cases, the stone foundations have been replaced or faced with concrete.
The piers of bridges with more than one span may have been built with the bridge and are probably of stone masonry, or they may be more recent additions to the structure. Piers added after the original construction may be timber or steel bents, or they may have been concrete structures. The bents often are made of piles (usually three or more, in clusters) driven beneath both trusses with a cap (horizontal member) connecting the tops of the piles. The cap might be directly beneath the bottom chords of the trusses, or additional blocking might be installed as a fill. These bents were added in reaction to perceived or visible weakness in the original structure. An example of a retrofit bent is shown in figure 68. These groups of old telephone poles were not especially sturdy, but helped support the Hamden Bridge in Delaware County, NY, for many years. They toppled easily when removed as part of the recent bridge rehabilitation.
Adding these bents was probably considered appropriate and viewed as a permanent feature. More recent attention to historic preservation often views such actions as a quick, nonpermanent, superficial fix to help restore the bridge's capacity, but these remedies are not usually considered an acceptable long-term solution. It must be recognized that, in addition to altering the historic and visual characteristics of the bridge, introducing the bent changes the behavior of the trusses and often causes major distress in the truss elements.
The vast majority of the original covered bridge abutments were constructed on a base of stones that served as a footing. The bottom of the abutment was dug to a depth below the streambed, with the water diverted or separated from the foundation pit. Then a base was built with plan dimension larger than the main portion of the abutment. The base was usually built with larger stones, or even concrete. Then the stem or breastwall of the abutment was constructed to be large enough to spread the load over an area of soil, resulting in a base pressure that the soil could resist without a slide, slip, or overturning failure.
At the sides of each abutment (upstream and downstream), wing walls were built to retain the soil of the approach embankment. The wings may be at right angles to the abutment stem, or they may be flared, depending on the builder's preferences and the geometry of the bridge with respect to the abutment. The original construction of the wing walls would have used the same material as the abutment stem (i.e., usually stone, also supported by spread footings). Figure 69 illustrates the difference between wing wall and abutment stem. The abutment stem is the portion directly beneath the timber structure. The wall to the left (old painted sheet piling) and the old stone wall to the top right serve as wing walls to retain the approach fill from spilling into the river. This is from Fitch's Bridge in Delaware County, NY, before its recent rehabilitation.
In many instances, the bottom layer in a spread footing would have been a layer of timbers, trees, etc., as a means of making a platform on a muddy bottom. As long as this timber was continuously underwater (or below the water table), it did not rot; these components are often found intact when abutments are replaced. Mud sill is the most common term for this type of initial layer.
In instances where the native soils at a bridge crossing were considered to be less stable, timber piles would be driven to support the weight of the abutment and bridge, vehicular live loading, and overturning forces of the earth pushing the abutment towards the stream. This piling usually consisted of peeled wooden poles, up to 300-375 mm (12-15 inches) in diameter and as much as 9.1-14.2 m (30-40 feet) long. The piles would have been driven at spacings as close as 1-1.2 m (3-4 ft) in plan view.
Modern foundation technology, applied to the same design, would use concrete exclusively in lieu of stone masonry; the piling could be timber, steel, or concrete.
Some abutments in more scour-prone areas have been protected by larger stones (termed rip-rap) along the face of the stream banks and directly in front of the abutment.
In many instances in which covered bridges survive with exposed stone masonry abutments, a layer of concrete has been installed over the stone, beneath the timber structure. This layer of concrete is referred to as a cap. The cap tends to knit the stone together and helps distribute the loads of the timber structure and vehicles over more of the abutment. It would usually have a vertical wall behind the horizontal bearing surface to keep the approach fill from spilling around the ends of the trusses. Often, by the time rehabilitation of the covered bridge superstructure is required, the added concrete cap also has deteriorated to such an extent as to require replacement.
The remaining stone masonry may contain cracked stones, perhaps even shifted stones, indicating failure of the stone against the lateral earth pressure. The cracked stones are often the result of differential settlement of the foundation or mud sill. Figure 70 presents an example of badly cracked stones in the bottom of an abutment stem. This condition precipitated removing the bridge from this abutment and completely replacing it.
In many cases, a concrete facing has been cast directly against the stonework. Depending on the details of the work, the stonework may either be completely hidden or left in some detectable form. In other cases, the joints of the stone masonry may have been pointed (filled at the surface) with mortar, the entire surface may have been parged (covered with a thin coating) with concrete, or the surface may have been coated with a thicker application by hand trowel or pneumatic equipment (often termed gunite). Figure 71 depicts a parged stone abutment-all joints have been filled in with a slurry mix of concrete, and the stone faces are still showing.
No matter what was used originally and subsequently added, trees and brush are often growing in the crevices of the abutment foundations, whether they are built of stone or consist of cracked and deteriorated concrete. Figure 72 demonstrates damage to a stone wall caused by tree roots; as the roots enlarge, they displace the stones. If the tree with such a poor root structure falls, it will dislodge a large area of stones. If the tree dies, the mass of rotting root structure will allow the stones to become dislodged. Left unchecked, trees will completely dislodge stonework and cause foundation unit failures. It is better to cut the trees than to pull them out, to avoid possibly dislodging the stones.
Another issue related to stone masonry foundations is the occasional significant loss of the finer backfill embankment material through the stone joints and crevices. Although the fine soil material will not necessarily fall out through the stonework of the abutment, it can be washed out during flooding. Large voids may result behind the stonework, increasing the risk of collapse, both in the approach roadway and/or the abutment itself.
An engineer preparing to rehabilitate a covered bridge, subject to the various situations and conditions introduced above, must evaluate various potential actions.
If the foundations are in suspect condition, based on the engineer's inspection, replacement may be an appropriate solution. This is a safe, albeit expensive way to address the uncertainty. Unfortunately, this action is often required after decades of neglect. However, frequently the decision to replace the stone is made without considering the ramifications of such a decision.
Concrete replacements may destroy a feature that contributes significantly to the historical and aesthetic flavor of the site. Repairing deteriorated stone foundations by relaying the stones with supplemental new stone is another acceptable action and often provides beautiful and historically interesting foundations. Either action is costly, yet good quality stone foundations can be more durable than concrete and often outlast the bridges themselves.
Therefore, one should always proceed carefully when evaluating existing foundations. Sometimes, replacement is absolutely necessary. However, as noted above in the discussion of problems and in the following paragraphs dealing with individual components, it may be acceptable to retain the existing masonry abutments, with some modifications.
Deteriorated concrete abutment caps usually can be replaced without damaging the underlying materials, whether they are stone or concrete. Admittedly, it is sometimes difficult to replace a cap in close proximity to the timber structure directly above. The superstructure might have to be jacked upward to provide room for such work.
Mortared joints or parged surfaces often trap moisture and actually cause more damage to the foundation by not allowing proper drainage of the embankment material. Therefore, exercise caution when considering automatically repointing the mortar joints or repairing the parged surfaces. It might be better to remove the parging and joint material to allow an evaluation of the underlying materials. If the joints have been mortared after original construction, or parging has been placed over the surface of the stone, it was probably introduced because it was considered necessary. If repointing and/or parging is to be retained, consider installing new weep holes (drainage openings) to allow drainage to pass through the abutment and wing walls.
Stone masonry that contains stones that have cracked in place is not, in itself, a reason to replace the stone. A single-span, timber-covered bridge can tolerate a fair amount of differential movement at the abutments without significant distress in the superstructure. Hence, an old abutment with stones that were cracked by weathering or some differential settlement has probably settled into a stable position, unless some other problem is causing ongoing movement of the foundation.
AASHTO requires a scour inspection of all bridges on a periodic basis and after major flood events. Therefore, bridge engineers need to remain vigilant against potential scour that might undermine bridge foundations on spread footings. A recent covered bridge rehabilitation project in East Delhi, NY, (see figure 73) included retaining the stone masonry abutments from 1859, but steel sheet piling was installed around three sides of both abutments to increase protection against scour action. Retaining the stone abutments was judged to be acceptable and resulted in considerable savings, compared to the cost to completely replace the abutments. Retention was considered a strong endorsement of the capabilities of stone abutments to continue to serve for a long time to come. A new light-colored concrete cap is visible immediately beneath the timber structure. The space between the old concrete and sheets has been filled with concrete. New large stone slope protection has been added to protect the slopes from scour.
Of course, what may be considered acceptable to some, in instances such as the example cited above and shown below, may not be acceptable to others. Chapter 18 of this manual discusses the Secretary of Interior standards for historic preservation and notes the judgmental nature of those standards. The combination of concrete, old stone masonry, and new stone masonry for slope protection is obviously a compromise and was strongly influenced by cost and practical solutions.
Figure 73. Modified original stone abutment-Fitch's Bridge, East Delhi, Delaware County, NY.
The timbers used to frame the load-bearing longitudinal trusses should not be supported directly on the foundation, whether they are concrete or stone. Any timber in direct contact with stone or concrete and in an intemperate environment will gradually deteriorate from the moisture that condenses on large masonry surfaces and the debris that inevitably accumulates on top of the foundations.
Accordingly, some sacrificial timbers should be installed; these occasionally can be replaced without incurring the major expense of replacing primary structural components of the trusses themselves. These buffering timbers are identified by a variety of terms, including bearing blocks, bedding timbers, and cribbing. Bearing blocks is the preferred term, because it depicts what the timbers are expected to do-they transmit the heavy weight of the bridge to the foundation.
The number, size, material, and arrangement of the bearing blocks vary, according to the preference of the engineer and/or contractor, but should be consistent with the type of truss being supported. For example, a Town lattice truss requires a larger (i.e., longer) area than a kingpost truss, because at least two intersections of lattice and chords should be supported at the abutment. The footprint size and distribution of blocks should be selected to avoid side grain crushing in the main timbers of the truss. The height of the blocks can vary widely, depending on the need to match an existing condition, or to fit the intent of new construction. The material is either hardwood or pressure-treated softwood. Figure 74 presents an example of a bearing block installation. This modified Burr arch support is from the recent rehabilitation of the Brown's River Bridge in Westford, VT.
The bearing blocks should be treated with wood preservatives before installation. Attempting to treat blocks in situ will not necessarily lead to long-term success. Although not authentic, some choose to use neoprene bearings instead of timber.
Another issue related to foundation conditions and bearing blocks is the fact that the top of approach roadways tends to rise over long periods of time; this is caused by deposits of granular materials at the entrance of the bridge, along with grit from snow control measures. This may indicate an opportunity during a bridge rehabilitation to raise the bridge to meet the new approach grade, rather than dig out the approach material. This can improve the hydraulic opening beneath the span and prevent approach drainage from entering the bridge. This can be accomplished by using thicker bearing blocks. Modifying the back wall of the abutment and top of the wing walls also will be required.
Some covered bridges contain special timbers between the bearing blocks and the bottom of the main trusses. Again, there are several terms: a common one is bolster beams. These large timbers are sized to extend longitudinally beneath, and usually directly against, the main timbers for a length of many feet. They usually project past the front face of the abutment for distances up to 3 m (10 ft) or more, although they sometimes are installed only over the width of the abutment. For those that extend only above the abutment, they may more appropriately be considered part of the bearing blocks.
Bolster beams are intended to shorten the bridge's span by extending the truss support beyond the edge of the abutment. Some believe that they are very helpful and important, while others disagree. Therefore, the size and length of these members vary greatly. These members raise the roadway surface above the top of the abutment. Their use may be precluded in a rehabilitation of a covered bridge that did not originally have them, and if the grade of the roadway will not be raised. Introducing them at a site that did not previously have them, without raising the grade of the bridge, introduces a new feature that may adversely affect the hydraulic opening beneath the bridge. An example of bolster beams beneath the Fuller Bridge in Montgomery, VT, before its recent major rehabilitation, is shown in figure 75.
Many covered bridges have been moved off their foundations by floods. Therefore, much effort has been spent in attempts to anchor these structures to their foundations. Although the details for such anchors vary greatly, they usually involve rods embedded in the abutments and bolted to the timber trusses. The efficacy of many such anchor attempts remains unproven.
A better solution is to locate the structure above the 100-year flood elevation. Admittedly, this involves raising the grade of the roadway and may require purchasing additional rights-of-way. If that is possible, the anchors may not be necessary unless some special event or unusually strong wind loading is envisioned. If the structure cannot be located above the 100-year flood elevation, then introducing a hold-down support may be prudent.
The hold-down devices should be sized with reserve to accommodate the inevitable section loss from corrosion that afflicts components on the tops or sides of foundations. The hold-downs can be anchored directly into concrete by initially placing them in cast concrete or grouting into drilled holes in existing concrete. In addition, they can be anchored in concrete that is placed in and around stone masonry (one must evaluate the potential uplift capacity of such installations). They also can be attached to drilled soil anchors. It is always wise to require a load test to verify the uplift capacity. A major weakness of many devices is the lack of restraint against side motion-vertical restraint is not necessarily enough. Figure 76 presents an example of a hold-down, although it was not complete at the time of the photograph. The galvanized rod projecting up from the concrete was completed with a heavy steel bar across the top of the bottom chord, to a mating rod on the opposite side of the chord.