A Design Error That Sat Undetected for Forty Years
On August 1, 2007, the I-35W highway bridge over the Mississippi River in Minneapolis collapsed during evening rush hour, killing 13 people and injuring 145. The National Transportation Safety Board's investigation traced the primary cause back to the bridge's original 1967 design — specifically, gusset plates at certain truss joints that were undersized relative to what the actual load calculations required, an error that had gone undetected through four decades of inspections.
Why the Error Went Unnoticed for So Long
Gusset plates — the flat steel plates connecting multiple truss members at a joint — are typically assumed by routine visual bridge inspections to be inherently stronger than the members they connect, so inspections historically focused heavily on the truss members themselves rather than independently re-verifying gusset plate capacity. The undersized plates had been adequate for the bridge's original design loads and had no visible distress for years, but the margin was thin enough that it couldn't absorb what came next.
If you want to see this logic applied to a real section, the Structyze Truss Gusset Plate Connection module runs the same checks live in your browser, free — useful for sanity-checking your own numbers in minutes rather than rebuilding a spreadsheet.
The Load That Finally Exceeded the Margin
At the time of the collapse, the bridge was carrying significantly more dead load than its original 1967 design — the deck had been widened and resurfaced multiple times over the decades (each resurfacing adding weight without necessarily being re-checked against the original gusset plate capacity), and at the moment of collapse, construction materials and equipment for an ongoing resurfacing project were staged directly on the bridge, concentrated near some of the already-marginal gusset plate locations. That additional load, on top of decades of accumulated dead-load increases, was enough to exceed the undersized plates' actual capacity, triggering the full truss span's collapse into the river.
Quick FAQ
Was the collapse caused by lack of maintenance?
No — the NTSB specifically identified a design error (undersized gusset plates from the original 1967 calculations) as the primary cause, compounded by accumulated dead load increases over the bridge's life and the construction loads staged on it at the time; it was not primarily a maintenance or corrosion failure.
Did this collapse change bridge inspection practice?
Yes — it led to a nationwide push in the US to specifically re-evaluate gusset plate capacity on older steel truss bridges as its own distinct inspection and load-rating item, rather than assuming gusset plates were inherently non-critical relative to the members they connect.
Final Thoughts
The I-35W collapse is a stark reminder that a connection detail's safety margin, once consumed by decades of incremental load increases nobody re-checked against it, can fail suddenly even though nothing about it looked different the day before. It remains the clearest modern argument for treating connection capacity as something to actively re-verify over a structure's life, not just assume from its original design.