Silver+Bridge+(Point+Pleasant)+Collapse

Heather L. Salasky, The Pennsylvania State University, Fall 2013, hzs124@psu.edu
 * The Collapse That Led To The National Bridge Inspection Standards **
 * Point Pleasant Bridge - December 15, 1967 **

Key Words
Collapse, Corrosion fatigue, Eyebar, Failure, Fracture, National Bridge Inspection Standards, Silver Bridge, S tress corrosion, Suspension bridge

Introduction
toc The Point Pleasant Bridge was designed by the J.E. Greiner Company and built by the Gallia County Ohio River Bridge Company to connect Point Pleasant, West Virginia to Gallipolis, Ohio (West Virginia Department of Transportation 2013, pp 1). It had a 22 foot roadway and one 5 foot sidewalk (LeRose 2013, pp 1). The two-lane eyebar suspension bridge took only one year to complete and was open to traffic on Memorial Day of 1928. The center 700-foot span was suspended over the Ohio River and the two 380-foot end spans were anchored on either side of the river (West Virginia Department of Transportation 2013, pp 1). The bridge was unique for a suspension bridge because each cable was made of a set of eyebars instead of the traditional wire strands, the eyebar chains made up the top chord of the stiffening truss over a portion of each of the spans, and each tower base had a rocker to allow it to rotate if the load on the bridge became unbalanced (Dicker 1972, pp 61). Local residents would come to call the bridge “The Silver Bridge” because of its silver colored aluminum paint (West Virginia Department of Transportation 2013, pp 1).



During the early evening of December 15, 1967, the Point Pleasant Bridge collapsed during a period of heavy traffic and 46 people were killed. Another nine people were injured. Of the 37 vehicles on the bridge, 31 fell during the collapse. Upon investigation, it was found that the collapse was due to a defective eyebar that experienced a cleavage fracture in the lower part of its head (Bosela et al 2013, pp 64). Since the eyebars were not designed to be redundant, failure in one eyebar would disrupt the continuity of the suspension system (Bosela et al 2013, pp 64). This disruption is what caused the bridge to collapse suddenly.

When the bridge was designed, stress corrosion and corrosion fatigue were not known to cause a problem in the type of bridge material being used for the Point Pleasant Bridge under the present exposure conditions (NTSB 1970, pp 9). The location of the flaw in the eyebar could not have been found via visual inspection, and moreover, the flaw could not have been discovered by any inspection methods available at the time without disassembling the entire eyebar joint (NTSB 1970, pp 9). This collapse would eventually change the way that the United States inspected bridges and lead to the development of the National Bridge Inspection Standards (Delatte 2009, pp 80).

Design and Construction
The Point Pleasant Bridge was intended to be an engineering marvel that would incorporate the newest technical innovations into the latest bridge design (Bullard et al 2012, pp 7). Designed during a time of rapidly changing technology, new ideas were always being integrated into engineering construction projects. The people of Point Pleasant looked forward to benefiting from the advantages of using new technology and were proud that their town could display a new type of construction (Bullard et al 2012, pp 7).

The bridge would be different from traditional suspension bridges since it used interlocking steel eyebars in series to carry the bridge deck instead of using traditional wire cables to support the deck. The Florianopolis Bridge in Brazil was the only other bridge built at the time that used this principle (Bullard et al 2012, pp 14). Later on, another similar bridge was built in St. Mary’s, West Virginia, about 90 miles north of Point Pleasant (Bullard et al 2012, pp 14).

The Point Pleasant Bridge was a design-bid-build project. The J.E. Greiner Company designed the suspension bridge. This involved drawing plans for the bridge and calculating stresses for the structure. It was designed for the American Society of Civil Engineers' H-15 loading (Ballard 1929, pp 997). The original design used a parallel-wire-cable suspension system, with an alternative option that used a seven-strand long-lay galvanized-steel wire rope for the suspension system (Lichtenstein 1993, pp 249). The design was estimated to cost $825,000 (Delatte 2009, pp 71). After the design was completed, bids were made on the project. The winning bid came in under budget from the American Bridge Company.

The American Bridge Company’s design called for an eyebar suspension bridge that utilized heat-treated eyebars and a rocking base (Delatte 2009, pp 71). The suspension chains would be used for both the main and the side spans (Bennett and Mindlin 1973, pp 152). The heat-treated eyebars were designed to have an ultimate strength of 105,000 pounds per square inch, an elastic limit of 75,000 pounds per square inch, and a maximum working stress of 50,000 pounds per square inch (Ballard 1929, pp 998). It is important to note that the material used to construct the eyebars had only become available for structural use a few years prior to the design of the bridge (Bennett and Mindlin 1973, pp 152). While traditional carbon steel has an elongation at failure of 18%, the steel used for the eyebars has an elongation at failure of only 5% (Dicker 1972, pp 62). The eyebars would be erected in pairs and their lengths would be dependent on their location in the chain (Ballard 1929, pp 998).

Because of the high design stresses present, the American Bridge Company insisted that using their new design would cost less than the J.E. Greiner Company's original design. It was also thought that using eyebar chains to support the bridge deck would have advantages over conventional designs. They provided stress calculations showing that their bridge design would meet the performance criteria shown on the J.E. Greiner Company's drawings (Lichtenstein 1993, pp 249). As it would later turn out, the advantages associated with using eyebar chains would be outweighed by the difficulty of inspecting and maintaining the bridge. Since the eyebars overlapped where they joined the next eyebars in the chain, corrosion could easily be hidden from plain view (Bullard et al 2012, pp 7). Figure 1 shows the Point Pleasant Bridge prior to the collapse, the Point Pleasant Bridge after the collapse, and a typical eyebar used in the design and construction of the Point Pleasant Bridge.

The bridge took one year to complete and was open to traffic on Memorial Day of 1928. The Point Pleasant Bridge soon became known as the “Gateway to the South” (Bullard et al 2012, pp 16). Prior to the existence of the bridge, no direct path connected the commercially important cities of Columbus, Ohio and Charleston, West Virginia. The construction of the Point Pleasant Bridge connected these cities, thereby changing the lives of everyone in the area.

Collapse
After some forty years of service, the Point Pleasant Bridge collapsed without warning (Lichtenstein 1993, pp 253). The collapse occurred during a period of heavy traffic on December 15, 1967 at 4:58pm. When one of the eyebars on the northern side of the bridge broke, the weight of the bridge was no longer supported. In rapid succession the chains, towers, and bridge deck all collapsed. The left of Figure 2 shows the missing spans of the Point Pleasant Bridge after the collapse. Note the intact railroad bridge nearby in the right portion of Figure 2.



In less than 60 seconds, the bridge plummeted into the Ohio River along with 31 vehicles. The collapse began in the side span on the Ohio side of the river. A joint separated and an eyebar failed. As a result, the Ohio tower fell. This in turn caused the West Virginia tower to fall. The center span broke into three parts and turned over (Lichtenstein 1993, pp 253).

Sixty-four people fell with the bridge. Of these, 44 died, 2 were never found, and 18 survived (Lichtenstein 1993, pp 253). About eighty percent of those who died drowned while the other twenty percent died from severe trauma (Bullard et al 2012, pp 31). At the time of the collapse, the water temperature was a chilly 44 degrees (Bullard et al 2012, pp 46). These cold temperatures made it hard for people in the water to survive. The sun set around 5:07pm, making it even harder to locate survivors in the water. Of all of the survivors, only five had been pulled from the river (Bullard et al 2012, pp 46). Figure 3 is a photograph of some of the wreckage of the collapse. Onlookers described the collapse saying “the bridge fell like a card deck” (LeRose 2013, pp 2).



Charlene Wood, who had started up the West Virginia approach when the bridge started to collapse, provided a shocking first-hand account of the collapse. She stated, “I felt a shaking of the bridge. I wasn’t going across so I threw my car in reverse. The shaking was so severe my car died, but it kept rolling back because of the incline. As I was watching in horror, the bridge was falling right in front of my eyes. It was like someone had lined up dominoes in a row, and gave them a push, and they all came falling down and there was a great big splash of water. I could see car lights flashing as they were tumbling into the water. The car in front of me went in. Then there was silence” (Bullard et al 2012, pp 44). Figure 4 is a summary of the experiences of survivors from vehicles that were on the bridge at the time of the collapse.



Investigation and Cause of Failure
After the Point Pleasant Bridge collapsed, there was a lot of speculation into the reason why it collapsed. Some people thought that inferior steel had been used in the construction of the bridge, while others thought that too many vehicles were on the bridge at once, and others even thought that someone sabotaged the bridge and caused it to collapse (Bullard et al 2012, pp 91). The National Bureau of Standards completed one of the most thorough incident investigations ever undertaken in the United States (Bullard et al 2012, pp 91). By the end of the investigation, all hypotheses were discarded except for one. Figure 5 displays a flow chart from the National Transportation Safety Board’s investigation report showing all areas that were investigated in order to determine the cause of the collapse. Even highly unlikely causes of collapse were analyzed in this extremely thorough investigation.



The cause of the collapse was directly related to the innovative design. The eyebar chains supporting the bridge deck were made up of long, twelve inch wide, and two inch thick steel beams that had circular “eyes” at each end (Bullard et al 2012, pp 92). Each chain had two eyebars in parallel. When chains met, the two eyebars in parallel were connected to the next two eyebars in parallel in the chain via a single pin (Bullard et al 2012, pp 93). At these connections, the heads of the eyebars overlapped such that large sections of the heads were hidden from plain view, thereby making it impossible to see these sections when the chains were in place. Figure 6 illustrates a typical detail of an eyebar chain and hanger connection.



The initial failure of the Point Pleasant Bridge was due to a cleavage fracture in the lower portion of the eye of eyebar 330 at joint C13N, which is the first eyebar chain joint west of the Ohio tower (NTSB 1970, pp 9). A ductile fracture in the upper portion of the same eyebar followed, causing the eyebar to separate from the chain (NTSB 1970, pp 9). As soon as eyebar 330 separated from joint C13N, the sister eyebar 33 slipped from the pin (NTSB 1970, pp 9). This caused the north chain to become separated. The collapse began on the Ohio side span and moved towards the West Virginia side.

The National Transportation Safety Board investigation team attributed the cause of the collapse to the cleavage fracture in eyebar 330 at joint C13N (NTSB 1970, pp 9). The fracture was due to the development of a flaw over the course of the life of the bridge resulting from stress corrosion and corrosion fatigue (NTSB 1970, pp 9). The eyebar experienced extremely high stress concentrations caused by friction at the pin and had high stress concentrations at corrosions pits. Corrosion had been caused by exposure to air pollutants, including hydrogen sulfide and sulfur dioxide (Maranian 2010, pp 5; Scheffey 1973, pp 79). When certain air pollutants come into contact with certain types of steel, the air pollutants can advance absorption of hydrogen into the steel, which in turn makes the steel more brittle and prone to cracking (Scheffey 1971, pp 44).

Unseen cracks had been present and growing for at least a few years until one crack in eyebar 330 became deep enough to fracture the eyebar. Eyebar 330 was approximately 55 feet long, 2 inches thick, and had a shank width of 12 inches. The eye at each end of the member had an 11.5 inch diameter hole and an 8 inch limb on either side of the hole (Bennett and Mindlin 1973, pp 152). The crack was approximately 1/8 inch long when the steel fractured completely (Bullard et al 2012, pp 91). This flaw was significant enough to fracture the remaining portion of the eye due to the high local stress and the low fracture toughness of the steel (Bennett and Mindlin 1973, pp 152; Hertzberg et al 2013, pp 664).

Some people wondered if eyebar 330 was somehow inferior to the other eyebars used in the chains since it had been the one to fail first. Investigators conducted laboratory tests in order to determine if the steel used for eyebar 330 was in any way different than the steel used for the other eyebars. It was found that the steel used was exactly the same as that of the other eyebars and that similar stress corrosion cracks had been developing in other eyebars that had not failed yet (Bullard et al 2012, pp 98). Even though numerous cracks were present, there was no way to detect them because of where they were located. Figures 7 and 8 are sketches of specimens studied during the investigation.



In depth laboratory tests were conducted to explore the cracks present in the eyebar that failed via scanning electron microscopy, electron probe microanalysis, and optical metallographic techniques (Ballard and Yakowitz 1970, pp 321). Based on the microstructure, surface topography, and sulfur gradient present in the eyebar, it was concluded that stress corrosion was the reason that the crack spread and that the eyebar eventually failed (Ballard and Yakowitz 1970, pp 321).

Figure 9 shows investigators examining part of the eyebar chain that they reconstructed in Henderson, West Virginia there they laid out and partially reassembled remains of the bridge. The eyebars on the right fit in parallel into the large slotted section on the left. The connecting pin is not present in this photograph. Even though the eyebars may seem large in this photograph, they were much smaller and thinner than other bridge components (Bullard et al 2012, pp 95). The investigators had the challenging job of determining which of the hundreds of fractures in the salvaged wreckage was linked with the collapse and not the recovery operation and then which of these fractures was the one that initiated the failure (Scheffey 1973, pp 77).



Prevention
It is unlikely that the collapse of the Point Pleasant Bridge could have been prevented. During the time period that the bridge was designed, engineers did not know that stress corrosion and corrosion fatigue could occur in the type of material used for the Point Pleasant Bridge and under the exposure conditions present on the site (NTSB 1970, pp 9). Bullard claims that when the Point Pleasant Bridge was built “it was constructed well within the building standards of the time” (Bullard et al 2012, pp 100). Even though potential design flaws may be apparent now, they were not apparent to the original designers who thought they had designed a safe bridge.

Moreover, the flaw that led to the collapse was not visible and the only way that the flaw could have been discovered with the technology available at the time would be to disassemble the eyebar joint (NTSB 1970, pp 9). Today's technology allows for nondestructive testing methods to help detect defects like this (Biezma and Schanack 2007, pp 403).

Luckily only two other bridges have ever been designed like this one. One of which, the St. Mary’s Bridge, was demolished in 1971. The Florianopolis Bridge, on the other hand, is still standing and open to pedestrians. It is slightly different from the Point Pleasant Bridge in that each segment of its suspension chain contains four eyebars (Bullard et al 2012, pp 100). It also differs in that its eyebar chains were used only in the main span whereas the Point Pleasant Bridge used eyebar chains in the side spans as well (Ballard 1929, pp 997).

Response to Collapse
Immediately after the collapse, traffic was blocked and emergency response teams looked for survivors. Local agencies and volunteers frantically worked to help victims. About 90 miles north of Point Pleasant, the St. Mary’s Bridge was shut down since it was of a similar design and needed to be inspected to ensure it would not collapse just like the Point Pleasant Bridge did (LeRose 2013, pp 4). As news of the collapse spread, specialized units rushed to the scene and recovery efforts became more organized. Shortly after, the US Army Corps of Engineers took command and a structured plan was established so that each responding agency could make maximum use of their capabilities (Bullard et al 2012, pp 51).

After helping survivors and recovering the vehicles from the river, the focus was on removing the structure from the river so that ships could resume travel. The river reopened on December 21 and most recovery work was done by the end of December. All of the site work was completed by the beginning of February. Pieces of the bridge were collected for later reassembly for investigation into the cause of the collapse. All debris needed to be located, recovered, and cataloged (Bullard et al 2012, pp 71). The pieces were sent 1.5 miles away to Henderson, West Virginia there they were laid out and partially reassembled. The effort was a success as a majority of the pieces were successfully recovered from the site. A Witness Group was created to obtain statements from survivors and eyewitnesses, a Bridge Design Review and History Group was created to check the original bridge design, and a Structural Analysis and Tests Group was created to study the remains of the bridge (Scheffey 1971, pp 41).

The bridge known as the “Gateway to the South” was no more (Bullard et al 2012, pp 16). The collapse of the bridge created a logistical problem for locals. Once again, there was no direct path to connect the commercially important cities of Columbus, Ohio and Charleston, West Virginia. The Point Pleasant Bridge was the only bridge that crossed the Ohio River in the area, with the closest bridges being 14 miles north at Mason, West Virginia and 41 miles south at Huntington, West Virginia (Bullard et al 2012, pp 117). Consequently, it was crucial for the community to get traffic going again and to find an efficient way to transport people across the river. The short-term solution was to operate a ferry service between Ohio and West Virginia (Bullard et al 2012, pp 115). The long-term solution was to build a new bridge in its place.

President Lyndon B. Johnson created the President's Task Force on Bridge Safety to investigate the cause of failure, to find funds to build a new bridge in its place, and to develop standards to ensure that this would not happen to other bridges in the United States (Delatte 2009, pp 74). The National Bridge Inspection Standards were created in response to the collapse (Delatte 2009, pp 80). Two years after the collapse of the Point Pleasant Bridge, the Silver Memorial Bridge was opened (Bullard et al 2012, pp 115).

Conclusion
The most common causes of bridge failure are structural and design deficiencies, corrosion, construction and supervision mistakes, accidental overload and impact, scour, and lack of maintenance or inspection (Biezma and Schanack 2007, pp 398). In general, the collapse of the Point Pleasant Bridge can be attributed to design deficiencies and lack of inspection. Because this bridge collapsed, the country has an increased knowledge of structural behavior and a new set of inspection standards. Figure 10 is a photograph of the local newspaper article published the day after the collapse of the Point Pleasant Bridge that is on display at the Point Pleasant River Museum.



Nineteen major roadway bridges have collapsed in the United States since 1900. Approximately fifty percent of them failed because ships collided into them or because major traffic accidents occurred on them. Other sources of bridge failures include river scour, shifting riverbeds, ice collisions, and earthquakes. Only five out of the nineteen bridges collapsed because of poor design or improper maintenance (Bullard et al 2012, pp 8). The fall of the Point Pleasant Bridge is one of these five and, moreover, ranks as the worst roadway bridge collapse in United States’ history (Bullard et al 2012, pp 7). This is due to the high number of casualties since the bridge was under heavy traffic when it fell, and since most of the vehicles ended up in the river and it took several minutes for rescue boats to arrive on the scene.

Due to the collapse of the Point Pleasant Bridge and the tragic losses associated with it, a nationwide inspection of bridges was undertaken. Many bridges were found to be in need of repair and received the maintenance work that they needed. Therefore, the repairs brought about by the collapse of the Point Pleasant Bridge prevented additional collapses from occurring in the United States. As Bullard eloquently states, “These reforms are the silver lining of the Silver Bridge disaster. While its collapse tragically ended the lives of many innocent victims, it also led to critical safety reforms that have undoubtedly saved the lives of many others” (Bullard et al 2012, pp 8).

Additional Resources and References

 * “Silver Bridge Tumbles, Toll, 7 Dead, 41 Missing.” Point Pleasant Register, December 16, 1967.**
 * Newspaper Article: This article was published in the local newspaper the day after the collapse of the Point Pleasant Bridge.