Restoration Philosophy for Covered Bridges, Old vs. New

By Arnold M. Graton, Jr.


As a historic restoration builder, I have noticed over the last few years a disturbing trend in covered bridge restoration. More and more often, local municipalities that own covered bridges are turning to state and federal governments for funding of restoration projects.


This is understandable as restoration projects are quite expensive and usually there is not enough room in the local budget to cover the costs of a bridge restoration. Local officials often feel that it would be irresponsible to not take advantage of seemingly “free money” or to go against the advice of higher government officials who urge them to accept their grant funds.


The main drawback to accepting state and federal funds is that it obligates the municipality to abide by standards that are set by the State and Federal Departments of Transportation (DOT). The standards they set for bridge repairs are formulated to address modern traffic and building practices. While these are fine for a new bridge, these standards often compromise the original aesthetics and traditional engineering of a covered bridge. Also, they seldom accommodate standards of historic restoration.


Current standards tend to impact a covered bridge restoration project in four ways:

road clearance, load capacity, road width, and wind load.


Let’s first address load capacity. Many times a State DOT will require that when a covered bridge is repaired it be upgraded to a much higher load rating than it was originally designed for. HS20 is popular as this is the standard for most highway bridges and qualifies the bridge to carry any non-permitted legal load. In most cases then, a covered bridge that was originally designed for a 6 or 12 ton load rating will require a substantial design change in order for it to be upgraded to a HS20 rating.


Accommodation of increased load is most often accomplished by beefing up and arguably, over building the floor system. Another popular way of increasing load capacity is to add new piers in the river and run steel stringers from pier to pier to carry the bridge deck. This takes the original bridge trusses out of play. The net result is that it turns the bridge into a romantic shelter sitting on a steel bridge rather than a functioning historic covered bridge.


The next issue is higher road clearances. In some cases, modern standards call for an increase in the height of the bridge to allow taller vehicles to transit the bridge. The reason often given for increasing a bridge’s height is to allow newer and larger emergency vehicles to use the bridge. This is the case with the Bath-Haverhill Bridge that crosses the Ammonoosuc River between Bath and Haverhill, NH. In this particular case, the solution proposed is to hang the floor system below the bottom cord.


An obvious drawback to this plan is that it changes its historic design of the bridge. Lowering the floor to increase the height would result in the loss of the floor system of the bridge, which is very early and perhaps original. Additional height would also encourage larger vehicles to use the bridge causing possible overloading problems.


The next issue is the width of a bridge. In cases where a historic bridge has been destroyed and the community rebuilds it, the trend is to increase the width of the bridge from a single lane to two lanes. The only way to accomplish this is with a much larger structure, usually at an astronomical increase in cost.


Widening the bridge also means that the original abutments can no longer be used. The original abutments are usually fine examples of historic craftsmanship and engineering. New abutments are costly and because they are being replaced anyway, there is a temptation for the highway department to straighten the road and approach to the bridge. This also adds to the cost and sometimes results in the bridge being relocated.


The last challenge relates to wind load. Ever since the suspension bridge known as “Galloping Gertie,” over the Tacoma Narrows in the state of Washington collapsed in 1940* wind load has emerged as a serious concern for engineers. Traditionally covered bridges have taken an intuitive approach to wind load. Now, however, the wind has become the subject of extensive math calculations and very high velocity assumptions which result in nontraditional solutions. I’ll provide an interesting example of this later in the talk.


A brief look at the history of two covered bridges, the Corbin Bridge in Newport, NH and the Smith Bridge in Plymouth, NH reveals two distinctly different approaches to the rebuilding of historic covered bridges. Both the Corbin Bridge and the Smith Bridge burned in 1993. In both cases the communities wanted to rebuild their covered bridges and our family firm, Graton Associates, gave quotes to both towns for the reconstruction of them.


The original Corbin Bridge was built in 1835** and remained in good working condition until it was destroyed by arson in 1993. The town of Newport was determined to have an authentic reconstruction of the Corbin Bridge and therefore, voted to rebuild their bridge without State money. The cost of replacing the bridge was covered by the insurance settlement. The New Hampshire Department of Transportation convinced the town to build the new bridge to HS20 standards. The community raised private donations to defray the additional cost, so that public funds would not have to be used.


The community of Newport became a large part of the whole process of rebuilding the bridge. Many of the residents came by every day to watch the progress. The bridge was pulled across the river using traditional methods including oxen to provide power. A local writer, Patrick O’Grady, worked on the bridge and produced a valuable publication documenting the entire project. The event culminated in a large community celebration. The best part was that the final cost of the bridge was exactly what had been quoted originally and no taxpayer funds were used.


The story of the Smith Bridge in Plymouth presents a different scenario. The original Smith Bridge was built in 1880** and by the 1970s it had fallen into serious disrepair. The bridge was finally closed to traffic in the early 1990s. In an effort to save the bridge, a citizen’s group launched a grass roots effort to raise the needed funds to restore it. Sadly, it was destroyed by arson in 1993 before the restoration could begin.


The Smith Bridge was insured so the town of Plymouth set out to solicit estimates for an insurance settlement on the bridge. We gave them a price based upon an exact replication. At this point the town had a choice to make rebuilding the bridge based upon the quotes received and the insurance settlement they would receive or accepting the State’s DOT’s proposal of “80/20” money. This program establishes a formula where the State pays 80% of a project’s cost and the town pays 20%.


But, as we all know, there is no such thing as “free money”! One of the requirements of receiving “80/20” money is that the DOT requires a town to build a bridge according to their specifications. The bridge was upgraded to HS20, widened to two lanes and the clearance was increased to 14 feet. The abutments were moved down stream about 50 feet and the road was straightened and widened. All of this greatly increased the cost of the project. In this case, the State offered to accept the insurance settlement as the required 20% and call it even. The 80%, of course, was covered by the taxpaying people of New Hampshire.


Thus, in Plymouth a 6-ton one-lane bridge with a low road clearance suitable for passenger cars and light trucks was replaced with a two-lane HS20 bridge with a 14-foot road clearance. Because the bridge was so large, the timbers had to be laminated in a factory out of state. Engineers, with no background in traditional timber framing, designed the bridge. DOT required all timbers to be treated in pentachlorophenal (a highly toxic substance that is illegal in the USA) resulting in unpleasant working conditions for the assembling crew and the bridge emits a noxious odor to this day.


The new Smith Bridge is four times the size of the original and the total cost to taxpayers was many times what it would have been to build an exact reproduction of the original. Also when dealing with government programs there is often a waiting list. Newport got their bridge back in 1995, but Plymouth had to wait until 2000.


My point is not that there is anything wrong with the new Smith Bridge. It is a beautiful monument to traditional design and building techniques. Additionally, it proves to skeptics that a covered bridge made entirely of wood can be built to handle modern traffic. But it is not a good example of historic reconstruction.


I feel that we can always improve on an old design but when we set out to restore a covered bridge our goal should not be to change the design of the bridge, rather it should be to return the bridge to its own optimal state.


Here’s an analogy--What if the Smithsonian asked you to restore the Wright Flyer to working order? The first thing you could say is “We really need this thing to fly farther and the reason it has such limited range is that it is under-powered.” So we would put in a bigger engine. Then you could say, “The air frame is just not strong enough to handle this much power.” So, we would beef up the frame. Then it would be “Canvas wings are inefficient and unsafe. It really should have an aluminum skin.” Well, before you know it, we’ve turned it into a Mig 21 instead of the Wright Flyer. Similarly, our goal should be to restore a bridge to its optimal condition, not change its design.


I would like to emphasize at this point, that when a state or municipality is thinking about restoring a covered bridge and is putting together a design team, it is very important to get a restorationist or a covered bridge specialist on board early in the process. Too often, a project goes out to bid with plans and specifications that have been put together by an engineering team that has little or no experience with covered bridges. Once the design work is done and a project goes out to bid, it is too late to change anything if the project is flawed. You just have to bid on the plans you get. If an experienced covered bridge restorationist is on the team from the beginning then you will save money and the finished product will be better.


An interesting example of efforts to upgrade an existing bridge is the Burkeville Bridge in Conway, MA. Our company Graton Associates gave the town of Conway a price for restoring it about 20 years ago but no action was taken. The bridge has been closed to traffic for more than ten years. Mass. Highways has taken over the project and they now want to restore the bridge and reopen it to foot traffic only.


They have two problems. First they are requiring a floor load rating of 40 lbs. per square foot. This means that as a footbridge it has to be heavier than it was for cars. Next and even more significant, is the issue of wind load. In every bridge project I have been involved with for the last seven years, wind load has been a concern. Computer modeling consistently shows that all traditional covered bridge trusses fail under wind loads required by modern specifications, even though there are examples of bridges all over the world that have been withstanding windload for a century or more.


From a historic restoration point of view, while it is appropriate to design a new bridge to stand up to 100 or 150 mile/hour wind, it is not appropriate or necessary to change the design of a historic bridge to accommodate modern specifications.

Over the last ten years Mass. Highways has solicited three designs from three engineering firms. Because all three proposals involved substantial design changes, including steel brackets to resist imaginary wind forces, the Mass. Department of Historical Resources has rejected all three proposals.


The company that tackled the problem most recently, contacted me to help them address the design issues using traditional methods. We were able to address the live load requirement for the deck without much trouble, but the wind load requirement was a different matter. The most frustrating aspect of this type of work is that modern day engineers feed all their data into computer programs that do not have the capacity for analyzing traditional wood joinery. Thus, the computer kept coming up with “moment” connections where none existed.


Consider, if you stack one wood block on top of another, the joint between them will be very strong in compression but will hold nothing in tension or “moment.” This same situation is very common in traditional covered bridge design, where joints are designed to work in only one direction. In some cases, a member that is designed to work in compression will simply fall out if it comes under any other force. This was the case in Conway.


I came up with several solutions to increase the ability of the bridge to withstand wind load, using traditional methods. Obviously, this involved design changes, but at least it was technology that would be recognized by the original builder. However, all attempts failed to hold the required wind load in the computer simulation. In the end, the only way the engineer was able to make the simulation work was to use steel brackets to create moment connections where there previously were none. However by doing this, we lost the craftsmanship, design, and technology that we were trying to preserve. Also this introduced new forces to members (mainly bending and tension) that they were not meant to handle. This, of course, was totally unacceptable and the plan was rejected.


In this case, having a covered bridge expert on the design team was a good idea but unfortunately, we could not overcome the obstacles presented by the engineering requirements. This is primarily because the wind load requirements are unreasonable and the computer program they were using could not properly analyze the truss design under wind load.


I am not concerned with truss analyses today but rather, in the requirements. If a bridge can stand in the wind while in bad disrepair then it should hold up that much better if repaired to its optimal state. This should be our goal. To repair the bridge to its optimum state and not to change the design.


I recently got a call from Jim Stafford who worked with me on the Gilbertville covered bridge in 1986 and 87. He told me that the Gilbertville bridge had been closed by Mass highways and asked me if I would come down to look at it. Though the bridge was in very good shape compared to most bridges I see, it did have some problems.


It is interesting that the primary reason they closed the bridge is the way they analyzed the deck. Joe Bianchi, the Engineer on the job when I repaired the bridge in 1986 explained to me that there is a range of values that you can give to the wooden members to analyze a deck system, and that the wood technologist the state hired had used the lowest end of the range. Because of this, an engineer could repair the bridge simply by using higher values for the wood members.


This brings up another interesting trend. The tendency to make the bridge decks as strong as or stronger than the bridge trusses. Traditionally the deck is the weak link in a bridge. Every system has to have a weak link -- it’s unavoidable. If the weak link in a bridge is in the truss and the bridge is overloaded the whole bridge is lost. If however, the deck is the weak link and the bridge is overloaded, the floor lets go and the offending load is released. The deck must be repaired but the bridge is preserved.


The reason for strengthening the decks is humanitarian. Human life is more valuable than a bridge. But, if you cause the truss to fail rather than the deck the overloaded vehicle is still in the river but now it is in a tangle of wreckage.


My point is that we should be flexible with using modern specifications on historic covered bridges. Whenever you increase the capacity of a bridge in any significant way you change the historic design and the restoration effort is compromised. No engineer or builder, regardless of how skilled can make a bridge hold more than it was originally designed for without changing it. You can’t haul gravel in a Ferrari.

It is my contention that we should be striving to preserve, not just the outward appearance of our covered bridges but also the design, workmanship and technology of the original builders. In this way thousands of years of accumulated knowledge can be preserved for future generations to learn from and enjoy.




** World Guide to Covered Bridges, The National Society for the Protection of Covered Bridges, 1989.


Draft: 4/30/03