A writer to the Post-Dispatch this morning suggested that in the 60s, China was selling steel to the US; I can't vouch for the assertion by the writer, however my question is for the bridge builders--if there had been a new supply of steel (say even from a different US supplier than you have used in the past), would it have been standard operating procedure to test the steel once or even periodically for various parameters like hardness, to make certain the steel meets project specifications? Would there have been periodic testing to ensure continuing quality?
DOTs have and use an extensive QA/QC process to verify the materials in their bridges. As far as I am aware, they require fabrication shops to be certified by the AISC in order to be considered qualified to fabricate for bridges. The AISC certification process further imposes requirements on their fabrication procedures and record keeping.
Steel supplied to DOTs for major structures is traceable back to the place it was melted into ingots. It is tracked every step of the way to the bridge. In the fab shops, each piece is marked to indicate its place in the bridge. The shop drawings provide the information to locate which sheet of steel, or structural shape, the piece came from.
The tracking numbers are called heat numbers. Each heat is tested for chemical composition and mechanical properties. When required, additional tests such as hardness, charpy V-notch impact, etc. are performed and supplied to the fab shops and often forwarded to the DOT (owner).
Most of this QA/QC process is directly or indirectly mandated by the FHWA because of Federal Funding. In addition, unless exceptions are granted, iron and steel products have to be made in America, start to finish. I hear there are two significant projects overseas, one being fabricated in China for the State of California. I do not know the "Buy American" status of that specific project.
Regarding the comment about bridge behavior with "frozen" bearings ...
In my area of the country, bridges are designed to withstand a seasonal temperature fluxuation of 120 degrees F. Using the coefficient of expansion of steel, this corresponds to a temperature strain of 0.000780, half up and half down. The problem can be considered as a prestress force analysis. If you preload the steel with a strain of 0.000390 and release it, what happens?
In a determinant structure, it expands or contracts by this strain. If it is indeterminant, multiply the strain by 29,000 ksi, and you get something near 11 ksi internal stress. The structure will seek a new equilibrium and much of this load will cause additional force in the supports. There will be some raising of the deck and some additional force in the piers. The biggest problem is quantifying the degree of restraint at the supports.
Dinosaur - your comments are appropriate for the state of the steel fabrication industry now, but in 1960 which is PROST's point, I know that AISC didn't certify fab shops and most states didn't have a "buy America" policy either.
Regards,
Qshake
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
Yes, I'm sorry I was interrupted while writing that long post and lost my train of thought. I had meant to mention the disaster in 1967 that caused many of these policies to be enacted. However, I believe steel was still tracked by heat number even before the Silver Bridge collapse. Many welding and materials policies were upgraded after that. Hopefully a similar process improvement can be obtained from this accident, but I would be surprised if we learned as much this time around. We still build non-redundant fracture critical steel structures and install pin-and-hangers in new construction. Sometimes I feel I am beating my head against a wall.
Perhaps there could have been Chinese steel in the bridge. In high school, we spent a full term studying Russia and China; I recall a lecture about the Chinse steel industry in the 50's & 60's: Complete disaster.
At the time this bridge was built US Steel and Bethlehem were still the major players in the US.
The only imported steel in any volume at that time was the Japanese steel and I do not think the volume was that great, much of that was on the west coast.
When I worked on a U.S. Steel plant under construction in Minnesota in 1966, the main concern was grinding the "Bethlehem" off some steel from the local fabricator when he let it get by on change order fabrications.
There are many problems with steel today because the demand is so high.Mills in China and India are sometimes selling one type of steel for another [200 for 300 stainless].Even inventing new grades [ 200 stainless].Out of spec steel. Steel with high amounts of trace elements. Best to deal only with reliable mills, demand documentation [chemistry, mechanical properties etc] even though more expensive !!
concretemasonry got it very close to right as both Bethlehem and USS both embossed there names on what they produced. Each mill had a slightly different marking. There was no particular effort to keep track of the material as specific heat numbers but there were run numbers that with a little effort one could get the actual chemistry and physicals of a run. During the Korean War a lot of steel shapes came from Europe.
The actual tracking of a particular heat of metal didn't gain momentum until Adm Hyman Rickhover required all materials used in the construction of Submarines be tracked. He came down hard on all of metal producers that supplied materials for the Navy. It was quite helpful as the metals industry went across the board. The biggest change was in welding consumables as you didn''t have to remember the color codes and wander why one batch performed better than another.
The Nashville Bridge Plant near by boyhood home always slipped some paint on the embossed name on the steel used in fabrication so it was quite visible when loaded on a rail car for shipment. At this plant it was TCI, the Tennessee Coal and Iron Co.
My brother has talked to one of the old fabricators that work in the Nashville Bridge Shop. They made components for steel bridges all over the country. The components they made were 99% hot riveted. All holes were drilled and reamed and if at all possible machine set using tools by Mesta. They only hole punching occurred during WWII for what he called light weight construction. All the connection holes were drilled and ream whether they were for bolts or rivets.
He commented that when the machine setters hit the rivets there was no problem with the rivet being tight. They only used American Steel.
Sidelight:
The Nashville Bridge site is being razed to build a modern DI pipe mill. If it hasn't been removed there is onsite a block of Zinc that is 6' wide x 8' deep x 110' long form the old galvanizing plant. Removing it is going to be chore as it contains about 2% lead.
In the last 20 years, we've used a lot of imported plate and domestic plate, and haven't seen a great deal of difference in quality- if anything, more problems with the domestic. Don't think we've seen much if any from China.
Last time I heard, several years ago, out of the ten largest steel mills in the world, none were in the US.
Regardless, if you build a structure and it operates fine for 40 years and then collapses under minimal load, that's not the kind of problem you'd normally associate with substandard materials.
My guess, because if you're not guessing, you're not playing along:
The compromise was with a connection from the secondary truss to the top chord of the main truss on the South and East side. The connection was corroded or fatigued or both; it was bad and the top chord of the truss that was supposed to be braced by the secondary truss was now only being braced by the decking. Contractor comes along to replace the decking, knowing what all contractors and bridge engineers know, that you can remove the decking because the bracing is provided by the structure... The top chord on the east (at the south end) buckles, splitting the deck first at the middle of the lanes and then pulling the entire south end to the east. From there the bridge collapses south to north.
Here is some additional information on bridges from another source:
Source: American Road and Transportation Builders Association
In view of extensive coverage following the Interstate 35-West steel bridge collapse in Minnesota, ARTBA released key highway and bridge statistics:
Of the 594,709 U.S. bridges, 152,945 (26 percent) are structurally deficient or functionally obsolete, according to 2006 Federal Highway Administration data. Of the 961,382 federal-aid road miles, 161,750 (17 percent) are reported to have conditions needing resurfacing or reconstruction.
Of Minnesota's 13,008 bridges, 1,586 (12.2 percent) are structurally deficient or functionally obsolete -- third-lowest in the nation, according to FHWA. Of Minnesota's 31,612 federal-aid road miles, 2,871 miles ( 9.1 percent) are reported to have conditions needing resurfacing or reconstruction.
According to the U.S. Department of Transportation's 2006 "Status of the Nation's Highways, Bridges, and Transit: Conditions & Performance" report: a) Structurally deficient means that significant load-carrying elements of the bridge are found to be in poor or worse condition due to deterioration and/or damage; or, the adequacy of the waterway opening provided by the bridge is determined to be extremely insufficient to the point of causing intolerable traffic interruptions. A deficient bridge, when left open to traffic, typically requires significant maintenance and repair to remain and service. And, b) Functionally obsolete bridges result from changes in traffic demand on the structure. For example, a bridge designed in the 1930s would have shoulder widths in conformance with the design standards of the 1930s. However, design standards may have changed since that time.
Not too technical, but another description of the terms being tossed around.
The video actually seems to indicate a main truss member or joint failure at the north end since it drops prior to the south end. Rotation, and lateral displacement could be caused if one main truss failed prior to the opposite main truss. I think in the end they will determine that one main truss joint failed, and lead to progressive collapse. I'm not a bridge engineer, but have designed some long spanned steel trusses.
I know this is a bit off topic, although Concretemasonry’s post made me think. Who is actually reviewing these bridges and roads? If it is a professional engineer reviewing a bridge that is “structurally deficient” don’t we as structural engineers have an obligation per our ethics to be sure that something is done about it. I have to wonder if I found a structure that was “structurally deficient” and only stated that it was structurally deficient in a report and it fell down and killed somebody, as a professional engineer where would I be in terms of adhering to my engineering ethics?
A source familiar with the investigation has told the Pioneer Press that investigators were looking at the possibility at least several plates, known as gussets, were too thin at half an inch. Just inside the piers lie four gussets labeled "U10." They're a half-inch thick, according to design records.
The video definiately shows that the failure occured near the south pier. The bottom chords near the north pier failed after the collapse, I think due to excessive rotation of the members during the fall. thereafter,the north approach collasped, I think due to excessive elongation of the top choars near the south pier. I also think, that the east and west side bottom chords near the north pier did not deform simultaneously, which may cause the entire span to rotate westward during fall. This is probably why the span was about 50' away from the south pier after the fall.
I truely believe that something went wrong near the south pier bearing.
It will be hard to fault the original design as it picked up two additional lanes of traffic.
Copied from ENR
"Engineers Swarm on US Bridges to Check for Flaws."
By Tom Ichniowski and Thomas F. Armistead, with Tom Sawyer, Jonathan Barnes, Bruce Buckley and Lucy Bodilly
"Meanwhile, in Minnesota, planners pushed ahead to have a replacement bridge constructed rapidly, although they quickly ran into conflicts over potential enhancements such as adding light rail and more lanes. The Minnesota Dept. of Transportation on Aug. 14 whipped out a preliminary design to solicit feedback. It features a 10-lane bridge with four shoulders, allowing room for expansion.The old bridge started as a six-lane crossing and later was expanded to eight by claiming most of the shoulders. The proposed alignment appears to match the old one."
I concur with imnotclever's guessings above.
Also adding that the wobbling experienced by the workers, that was increasing as the work (removal of concrete deck) proceeded, cannot be explained by the failure of some gusset plates, unless this failure occurred many days before the collapse. The wobbling should be caused by the buckling of a main element, that, till the collapse, was stopped and recovered by the intervention of some other structural restraint.
Wobbling - somewhere I read that all of the workers had been interviewed and none of them experienced anything abnormal [such as wobbling] before collapse !!
![[pipe]](/data/assets/smilies/pipe.gif "[pipe] [pipe]")
