Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: AP: Wash. State bridge collapses into Skagit River; sends vehicles, people into water

[Subject Prev][Subject Next][Thread Prev][Thread Next]

In this case of course the initiating cause of collapse was the impact of truck hitting the sway braces. But not all bridges being hit by a force that they are not designed for should totally collapse. Especially due to inherit ductility of steel as a material, steel structures can take quite a lot of beating before you can actually collapse the entire span. But, there is very important exception to this and that is all bets on ductility and redundancy are off if the bridge is "fracture critical" which this bridge was. In other words, it would have made any difference if this bridge was opened to traffic the day before yesterday and its rating of course would be 9/9 at the time of opening. Still because it was 'fracture critical" if it was hit by the same truck , it would collapse the same way. As you know, "Fracture Critical" bridge is a bridge that if a single member or connection fails (fractures, buckles or yields) the entire bridge can collapse. essentially fracture critical trusses are trusses that are determinate. As we know from our Static course, in a determinate truss you can find member forces by using only equilibrium, which is nice and you really do not need computers, but, if you remove a member or a connection from a determinate structure , it becomes unstable and collapses. That is what happened last night and what happened on August 1, 2007 in Minneapolis to I-35 bridge , which in the I 35 case resulted in 13 death and 112 injuries.

When in 2007, I explained to reporters why I35 bridge collapsed due to failure of single connection, I explained to them the meaning of 'fracture critical' since the bridge was listed as fracture critical and they wanted to know what it means, some in steel industry felt that I should not have talked to reporters, and said anything negative about steel structures. The Modern Steel Structures even had a pretty negative editorial on my comments, but, of course we have no choice other than telling the truth even if it is quite inconvenient.

I want to share also one thing with you and other friends on this forum and that is , one week after the collapse of I-35, a reporter from Minnesota called me and on the phone was asking me do I know what might have caused the collapse. I told him I cannot say much unless at least i have a chance to review drawings, and he said that the drawings are on the Internet, so, while he is on the phone I go to MnDOT site and look at the drawings and go direct to gusset plate sheets, since I had seen photos of a gusset sticking out in the water and clearly was fractured through the net section while it was up and not after it was dropped. So, I look at gusset plates and suddenly realize that the Gusset Plate U10 is only 1/2" thick Gr 50. Bridge details have the area and grade of steel on member. So, while this reporter is on the line, I do a calculation of tension on the vertical net section of the gusset plate and compare it to the strength and it turns out that the applied force is about twice as much as the net section capacity! So, I tell him that it seems to me that this Gusset Plate U10 should have been 1 inch but it is 1/2 and it is fractured and sticking out from the water. I added that this could be a possible cause although I need to dig in more . I check other gussets and sure enough all were fine. He prints my quote in the newspaper that 'Gusset plate U10 seems to be under designed and instead of 1/2 inch it should have been 1 inch". The issue goes back and forth, then 1.5 years later , my phone was ringing off the hook and it turned out that the final report of NTSB studies had been released that morning and it has concluded that 'Gusset plate U10 should have been 1 inch instead of 1/2 inch"

Today, so far I had 14-15 interviews with the press on the collapse of I-5 bridge and almost all of them were asking why a truck hitting a bridge totally collapsed it? Of course you cannot explain it unless you discuss the "fracture critical" nature of a small percentage of our bridges that makes them vulnerable to loss of a member or a connection In mid 1800s there were papers and discussions in bridge journals on whether trusses should be determinate or indeterminate. the final consensus was that they should be determinate (with one diagonal in each panel only) and main reason given was that if we have X diagonals we cannot establish the forces in diagonals. They said that X diagonals can only be used in the mid-panel where the shear force is zero and both diagonals have zero load.

Now , with computers at our disposal, do we need to design and build more "fracture critical" trusses? Why? Just we have been doing it so long , most likely even without knowing why we are doing it and what re the consequences?

I hope this time, instead of getting an editorial in Modern Steel Construction on personally attacking me it will open a debate and discussion on why do we need to design and construct "fracture critical" bridge trusses?

As for your suggestion to have ratings posted, there is no need for that. All bridges are rated and if a bridge is unsafe for use for intended purposes it will be closed or rated to carry lighter load. What we need to do to prevent a collapse like last night's, which could have been tragic , is to design systems that are not fracture critical.

With best wishes as always


On 5/24/2013 5:11 PM, bill(--nospam--at) wrote:
Hassan -

Thank you for the interesting and thorough explanation. I realize your
response is general in nature and I certainly understand the reason the
Washington bridge failed is due to the fault of the truck driver, but I
invite you to look at the pictures of the bridge that failed even in areas that did not yield. Also, the condition reports of 6 out of 9 etc bothers me. Maybe we should have the rating system displayed on the bridge like restaurants so that we as informed commuters can chose our route not only on convenience but risk as well.

T. William (Bill) Allen, S.E.
Consulting Structural Engineers
San Juan Capistrano, CA 92675
V (949) 248-8588/ F (949) 209-2509
-----Original Message-----
From: Abolhassan Astaneh [mailto:astaneh(--nospam--at)]
Sent: Friday, May 24, 2013 04:24 PM
To: seaint-seaosc(--nospam--at)

Truncated 1734 characters in the previous message to save energy.

Post your message to the list by sending it to: SEAINT-SEAOSC(--nospam--at)

The email messages sent to the list will be saved in an archive on the World Wide Web.
These archives are located at:

To contact the list owner, send your message to:

Sponsored By: Pacific Structural & Forensic Engineers Group, Inc. (PSFEG)

To unsubscribe, switch to/from digest, get on/off vacation, or change your email address, click here.