I+35+W+Minneapolis

toc Interstate 35W Minneapolis Bridge (Aug. 1, 2007) // Benjamin R. Barben, BAE/MAE, Penn State, 2009 //

Introduction
The Minnesota Department of Transportation (MnDOT) Bridge No. 9340, also known as the I-35W Bridge (shown in figure 1) was designed by Sverdrup & Parcel and Associates, Inc. of St. Louis, Missouri. The design was developed over several years and final design plans were certified by Sverdrup & Parcel and Associates, Inc. on March 4, 1965. The bridge design plans were approved by MnDOT on June 18, 1965. The I-35W Bridge over the Mississippi River was opened to local traffic in 1967. The bridge was built by Hurcon, Inc. and erected and staged by the Industrial Construction Division of Allied Structural Steel Company [1]. The structure used welded built-up steel beams for girders and truss members, with riveted and bolted connections [1],[2].

The bridge had thirteen reinforced concrete piers and fourteen spans which totaled 1,907 feet in length and carried eight lanes of traffic, four northbound and four southbound [1], [2]. Of the fourteen spans, the three main spans consisted of a pair of continuous steel deck trusses that spanned approximately 1,000 feet. Each span had two approach spans of 265 feet and a central span of 456 feet. At the north and south end, the two main trusses had 38 feet cantilever spans with 27 perpendicular deck trusses spaced at 38 feet. The deck truss cantilevered approximately 16 feet out past the main truss. Four piers supported the entire deck truss which was 108 feet wide [2]. The eleven north and south spans were a combination of steel multi-girder and concrete voided slab [3].

The bridge was carrying an average daily traffic of 141,000 vehicles in 2004. The average daily traffic of heavy commercial vehicles was 5,640 [3]. The earliest available average daily traffic figures for the bridge were from 1976 and an average daily traffic at that time of 60,600 vehicles [1].

Maintenance and inspection
The I-35W Bridge was routinely inspected once it was opened to interstate traffic in 1971 and had in-depth inspections annually, beginning in 1994. The major inspection findings included riding surface deterioration, deck joint failures, bearing corrosion, concrete pier deterioration, and approach span fatigue cracking [3]. In 1990, and for the following seventeen years, the NBI assessed the I-35W Bridge and it was given a rating of "poor condition." The NBI Standards provide a numeric rating system that rates sections of a bridge; deck, superstructure, and substructure on a scale of 0 though 9, with 9 being "excellent condition" and 0 being "failed condition." MnDOT has adopted a goal that no more than 2% of the State's bridges will be rated as "poor." In the Metro District, where I-35W is located, slightly over 6% of the bridges had a "poor "rating in 2005. In 1990, the I-35W Bridge's superstructure was lowered from 7 "good condition" to 4 "poor condition." The conditions associated with a 4 rating are "Superstructure has advanced deterioration. Members may be significantly bent or misaligned. Connection failure may be imminent. Bearings may be severely restricted." In respect to steel superstructures, "Significant section loss in critical stress areas. Un-arrested fatigue cracks exist that may likely propagate into critical stress areas [5]." A special inspection was carried out in May 2007 to inspect select deck truss members [3].

The I-35W Bridge underwent three major renovation/modification projects, two of which increased the dead load on the structure. In 1977, a 2-inch concrete overlay was added to the bridge deck to combat rebar corrosion. The additional applied concrete increased the dead load of the bridge by more than 3 million pounds. The 1998 renovation involved replacing the median barrier, upgrading outside concrete traffic railings, improving drainage, repairing the concrete slab and piers, retrofitting cross girders, replacing bolts, and installing an anti-icing system. The most recent major renovation was in 2007, which was to remove and replace the 2 inch concrete overlay added in the 1977 renovation [1].

Concerns over fatigue cracking in the non-redundant deck truss led to a research study by the University of Minnesota which was completed in 2001. This detailed assessment found: “The approach spans have exhibited several fatigue problems; primarily due to unanticipated out-of-plane distortion of the girders. Although fatigue cracking has not occurred in the deck truss, it has many poor fatigue details on the main truss and floor truss systems. Concern about fatigue cracking in the deck truss is heightened by a lack of redundancy in the main truss system. The detailed fatigue assessment in this report shows that fatigue cracking of the deck truss is not likely. Therefore, replacement of this bridge, and the associated very high cost, may be deferred [4]."

In December 2006, a steel reinforcement project was planned for the bridge to add redundancy. It was determined that drilling for the retrofitting would weaken the bridge, so the project was canceled in January 2007 in favor of periodic safety inspections. Internal MnDOT documents reviewed by the Star Tribune reveal that bridge officials talked openly about the possibility of the bridge collapsing and worried that it might have to be condemned [6].

Collapse
The final renovation project on the I-35W Bridge began in June 2007 and had a scheduled substantial completion date of September 21, 2007, with final completion expected by October 26, 2007. Progressive Contractors, Inc. (PCI), of St. Michael, Minnesota had completed seven pavement section overlays prior to August 2007 [1]. On August 1, 2007 the two outside northbound lanes and two inside southbound lanes were closed to traffic. PCI was preparing for the eighth section overlay, on the two inside southbound lanes. Materials and equipment were stockpiled on the bridge near the south end of the project [3].

At approximately 6:00 pm, during rush hour traffic, the I-35W Bridge experienced a catastrophic failure in the main span of the deck truss (see figure 2). 1,000 feet of the deck truss collapsed, with about 456 feet of the main span falling 108 feet into the 15-foot-deep river. A total of 111 vehicles were on the portion of the collapsed bridge. As a result of the bridge collapse, thirteen people died, and 145 people were injured. At the same time, a surveillance video camera just west of the I-35W Bridge, recorded a portion of the collapse sequence. The video showed the bridge center span separating from the rest of the bridge and falling into the river [1].

**The Gusset Plate Theory**
Following the collapse, the National Transportation Safety Board immediately began a comprehensive investigation. Approximately fifteen months after the collapse, the NTSB determined that the probable cause of the I-35W Bridge collapse was a lateral shifting of the upper end of the L9/U10W diagonal member which caused the gusset plates at the U10 nodes (shown in figure 3 & 4), which had inadequate load capacity for the original as-designed condition, to fail due to a design error by Sverdrup & Parcel and Associates, Inc. The U10 gusset plate failed under the combination of increases in the weight of the bridge, which resulted from previous renovations, and the traffic and concentrated construction loads on the bridge on the day of the collapse [1].
 * NTSB**

In addition, through examination of the collapsed structure, finite element analysis, and the video recording of the collapse, it was determined by NTSB that corrosion damage found on the gusset plates at the L11 nodes and elsewhere, the fracture of a floor truss, preexisting cracking in the bridge deck truss or approach spans, temperature effects, or shifting of the piers were not the cause and did not contribute to the collapse [1]. Additional findings include:


 * 1) Current Federal and State design review procedures are inadequate to detect design errors in bridges.
 * 2) Had American Association of State Highway and Transportation Officials, AASHTO, guidance included gusset plates in load ratings, there would have been multiple opportunities to detect the inadequate capacity of the U10 gusset plates of the I-35W bridge deck truss.
 * 3) The NBI Standards inspections would not have been expected to detect design errors.
 * 4) The conditions responsible for the Structurally Deficient rating did not cause or contribute to the collapse of the bridge.

Immediately following the collapse Governor Tim Pawlenty and MnDOT announced that the engineering firm of Wiss, Janney, Elstner Associates, Inc. had also been selected to provide essential analysis that would parallel the investigation being conducted by the NTSB. In November of 2008, WJE determined that the I-35W Bridge collapse began with the improperly designed gusset plates at a U10 node, in the vicinity of the L9-U10 compression diagonal. Failure at this node was followed very quickly by a similar failure at the other U10 node, which then led to the complete collapse of the deck truss system [3]. Additional findings include:
 * WJE**


 * 1) As designed, the deck truss members themselves were capable of carrying the loading during the time of collapse, with an appropriate safety factor.
 * 2) The as-built capacities of the gusset plates at the U10, U10', L11 and L11' nodes of the main span were roughly half of the capacities required by the governing design code.
 * 3) U10 and U10' gusset plates had not experienced any significant damage or deterioration, prior to the collapse.
 * 4) It was also found that some gusset plates were bowing before the collapse and WJE concluded that properly designed plates, with such bowing, would have been able to sustain the loading at the time of the collapse with an appropriate factor of safety.

**The Beam Theory**
Thornton Tomasetti was contracted to investigate the collapse on behalf of a consortium of 20 Minnesota law firms representing the victims and their families. Studies included assessing the impact of fatigue, temperature and load redistributions on critical bridge members subjected to static and dynamic loads [7]. According to Star Tribune, Thornton Tomasetti claims the cause of the disaster was not the undersized gusset plates but a failure of a nearby beam, the L9-11 chord, which is located on the lower part of the steel truss. This beam was not directly connected to gusset plate U10, but three other beams connected it to the plate. Thornton Tomasetti concluded that the beam buckled because of the hot weather and the loads placed on the bridge's deck the day of its collapse. It was also stated that the bridge's bearings, designed to allow for expansion and contraction, were corroded and kept the bridge from expanding in the summer heat, causing the beam to fail [8].
 * Thornton Tomasetti**

Due to current litigation of the I-35W failure case, the different theories have not yet been resolved.

Recommendations
As a result of its investigation of the collapse of the I-35W Bridge, the National Transportation Safety Board made the following safety recommendations[1]:


 * To the Federal Highway Administration:**
 * 1) Develop and implement, in conjunction with the AASHTO, a bridge design quality assurance/quality control program.
 * 2) Require that bridge owners use nondestructive evaluation to assess gusset plate condition of the truss bridges in their inventories.
 * 3) Update training courses to address inspection techniques and conditions specific to gusset plates.
 * 4) Revise reference material and address any newly developed gusset plate condition ratings in the AASHTO commonly recognized (CoRe) structural elements.


 * To the AASHTO:**
 * 1) Work with the FHWA to develop and implement a bridge design quality assurance/quality control program.
 * 2) Revise AASHTO Manual for Bridge Evaluation to include guidance for conducting load ratings on new bridges before they are placed in service.
 * 3) Modify the Manual for Bridge Evaluation to include evaluating the capacity of gusset plates as part of the load rating calculations performed for non-load-path redundant steel truss bridges.
 * 4) Develop specifications and guidelines to ensure that construction loads and materials placed on a structure do not overload the structural members or their connections during construction or maintenance projects.
 * 5) Revise the AASHTO Guide for Commonly Recognized (CoRe) Structural Elements to incorporate guidance for potentially damaging condition in gusset plates.