Readers give their view on why I35W Bridge Collapsed

It appears to me from the photos I have seen that the main north(?) pier is tilted appreciably towards the center of the river. So, was this pier pushed sideways during the collapse or did it tilt due to scouring and contribute to the bridge failure?

On 8/9 I commented about cyclic stresses and bridge resonance. Since that last entry, reports indicated nearly 600,000 lbs was present on the structure, prior to collapse. That weight focused in a relatively concentrated area would be another major factor. Plus jackhammering of the old concrete surface was also occuring - enter shaking forces! I still stand by a theory that low frequency, natural bridge truss resonance, induced by combinations of these various driving forces led to the collapse.

I would have to agree with the several engineers who have commented on the poor design of the bridge and the fact that there was not redundancy in its design. This is typical of the cost cutting civil engineering projects done in the 1960s. One only has to take a look at the Golden Gate or the Brooklyn Bridge to see that it is possible to built a safe sturdy design that, when maitained, will last indefinitely.

It may interest you to know that it has been said that there was several earthquake tremors about the time of the bridge collapse and there is a fault running in the water under the 35W bridge

The primary reason the steel members broke is metal fatigue caused by cyclic tension loads above the current design standard threshold. The metal fatigue causes cracks to form, and once a crack has propagated to a critical point, the member just snaps. The reason the bridge collapsed due to a broken member is there was no redundancy built into the structure. Bridges built today don't have this problem because the codes have been updated to prevent it. It is not easy to find fatigue damage. You need to know exactly where to look, and many times the fatigue damaged area is not visible or easily examined with testing methods, such as ultrasound or dye penetrant.

Reports of lateral swaying of the bridge in the days or weeks prior to the collapse sounds like the diagonal sway braces might have failed at the south end of the bridge. Reports indicate that these sway braces had corroded in the center, below the median. The 1996 report actually shows a photograph of a rust hole in the center of one of these sway braces. The salt spray system probably contributed to this corrosion in the center. Fatigue cracks will grow much faster in corroded members. Extra weight on the south end of the bridge would have contributed to lowering the natural frequency. Also, if the road surface had been weakened by the construction, it could have reduced it’s shear stiffening contribution. Side-to-side motion of this span might have cause a failure of the main truss over or near the south pier. Because this is a cantilever bridge, the center span would not actually be self supporting without the counter-balance moment of the two “wings” of the south and north spans. (If a bridge span is designed to be self-supporting between two piers, the truss would be deeper in the center. Since this is a cantilever design, the truss is deepest over the piers and thinner in the center.) The video shows a failure of the diagonals at the north end. But the video does not actually show the south end. If a failure had occurred at or near the south end, the north end would not have been strong enough to hold the moment of the whole bridge now cantilevered from the north end, and would immediately fail, which is what the video shows. The south end being the only section to collapse to the side is a big clue. I think this year’s inspection had been delayed by the construction. Sad.

Isn't it obvious? The pier on the south end is tipped into the river. The frame that was standing on and presumably bolted to that pier can be seen to be relatively undistorted. This would indicate the tipping of the pier was likely due to erosion undermining the footing of the pier and not due to stresses induced by the collapse. Stresses sufficient to push that pier over would have buckled the steel frame attached to it. I expect the investigation will show the pier tipped due to erosion, the attachments were sheared and the bridge stood, without proper support from that pier, for a considerable time. This would explain why the resurfacing crew members sensed swaying of the bridge. Ultimately, the bridge probably withstood the strain of that missing foundation longer than anyone had a right to expect.

I am a retired ME, but not in the civil engineering field. Mr. Martens'' overview appears to have much merit, especially the bridge load at the failure moment. Recent inspection reports indicated stress cracks and severely rusted areas. My thoughts center around the degree of rust, particularly internal rivet/bolt body integity. What inspection method (X-Ray?) can reveal this - at girder strut connecting plates, etc.? The fact of a 50-ft lateral motion on the south end at time of failure has to be a major key. Any one truss member letting go would apparently initiate a full failure mode. Plus include the millions of vibratory cycles of tension- compression-torsion over 40-years has to cause a shift in the bridge structure natural resonant frequency. These are all probable inputs that initiated the failure.

As mentioned in other postings, the road deck is usually an independent structure and the removal of the surface or the reinforcing bar would not contribute to the failure of the bridge support structure. The decking under the road surface is usually part of the structure and this might have been mistakenly compromised by the insertion of the de-icing sprinkler heads a couple years prior. This decking usually provides or contributes to the tensile, lateral, torsion, and sheer strength of the bridge. The overall design of the structure is also very poor as the steel trusses connect to the bearing piles some 20 feet in the air where there is no lateral bracing other than what is provided by the entire bridge as a whole. This design lacks structural redundancy which means that it was designed to fail at the first structural compromise, whatever that may have been. In this case it seems that there were a large number of extremely large semi-trailer cement trucks on the bridge at the time of the collapse as can be seen in the DOT traffic camera. You can also see one in the middle of the river where I believe the bridge initially failed based on the Army Corps video.

Failure of the bridge support system due to wash-out of the support structure in the river.

I believe there was a simultaneous failure in the concrete structure and the steel reinforcing bar. The concrete was probably already failing and an accident waiting to happen. According to Tessla's theories on the destructive effect of resonance, the steel was weakened by 30 yr. of resonance resulting from the continuous and heavy traffic. The steel finally became brittle and broke from the resonance, possibly like when a crystal glass can shatter from a shrill sound. I'm not sure if resonance effects concrete structures. There are bridge evaluation teams that can better diagnose these weaknesses. Regards, Pat Wilks

Dear Editor: Thank you so very much for such an informative article on the I-35W Bridge and its' collapse on August 1st. It is a very well written article that you have supported with a substantial amount of detail. It will give us all something to think about when we engineer our next project, especially when it comes to something like this. I am not an engineer, but I would have to say that one contributing factor to the bridge's ultimate colapse, would have to be something to do with the train passing under the bridge. Train tracks in any situation can cause vibration to the foundation and structure of any structure. That along with time, traffic flow on the bridge, and the repaving of the bridge could possibly have caused the ultimate collapse. How many times has the bridge been repaved? How much weight has been added to the structure each time it has been repaved? Has there been any new sewer lines added to the area surrounding the road that may have caused a stress fracture in the foundations under the bridge? A lot of unanswered questions that we will all get answered once the investigation takes place. Again, thank you for such a well written article. Korte Bros. Inc. Frank Morton