I35W Bridge Gussets: Who Made the half-inch Decision?

When the bridge was constructed, the center section of the bridge most probably arrived by barge via the Mississippi River. The center section of the bridge is lifted into place and attached to the two sections cantilevered out from the river piers. (See NYTimes photos -I 35W Bridge Collapse www.nytimes.com - photo 2) All the truss members meeting at the connection nodes may not be in the same plane. The iron workers erecting the truss must force the connection plates into place using tapered pins in the rivet or bolt holes. You will note that the rivet heads appear tight to the plate and it is a simple matter to determine whether the plate is tight to the truss member. The purpose of the connection plates is to transfer the load from the truss member to the rivets and then to the next truss member. Warpage as shown in the picture dated 2003 would have no effect in the transfer of the loads. The weight of the concrete trucks is a legal load. Meaning, the truck can be placed anywhere, anytime and any number of those trucks. In fact, the design load of any Interstate system bridge is for bumper-to-bumper Sherman tanks in all lanes. At the time of the collapse the actual load on the bridge was considerably less than the design load. Bill Kallman, PE www.myspace.com/billkallmanpe

The connections between truss members of Bridge 9340 were appropriately designed. The number of rivets and the thickness of the gusset plates were adequate to transmit any possible load between any of the members due to traffic, wind, etc. The focus of the state’s investigations was always the welded box members of the trusses. Measurements were made under traffic loading. The measurements showed that fatigue was impossible because the stresses were so low. The URS preliminary report said that total collapse under traffic loading was impossible unless many of the truss members cracked simultaneously - a very unlikely event. Based on my extensive experience with inspections of major truss bridges, the bowed connection plates indicate a slight misalignment in the diagonals of the truss at U10, which probably was there from the beginning. I continue to believe that the only force capable of collapsing the bridge was the pavement expansion force generated by the approach pavement. When the concrete deck of the bridge was partially removed, the absence of that deck material created a weakness in resisting that pavement expansion force. The long, slender compression members of the truss began to gyrate and the connections tore apart, releasing the center span of the bridge. If a section of the approach pavement had been removed and replaced with asphalt, the collapse would not have occurred. Consequently, the bridge in St Cloud Rt. 23 over the Mississippi River, could be similarly protected from collapse.