I-5 Bridge Collapse

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Blackwolf

Conductor
Joined
Nov 12, 2011
Messages
1,517
Location
CIC
Breaking News from Washington State:

I-5 bridge over Skagit River collapses (KING-TV, Seattle)

People and cars are confirmed as being in the water. This, if I recall, is just South of Bellingham, near Mt. Vernon. One whole span of the truss bridge has collapsed, and both directions of I-5 are affected.
 
I-5 bridge collapses over Skagit River; possibly triggered by truck
An Interstate 5 bridge collapsed into the Skagit River at around 7 p.m. Thursday evening, dumping at least two vehicles in the waters north of Mount Vernon.


A law-enforcement source said investigators believe a truck with an over-sized load heading south struck the bridge, which started bouncing, then fell.

The source said that 150 yards of the interstate dropped, sending a car, a truck and a travel trailer 120 feet down to the water.
...
All three people in those vehicles were rescued. They suffered minor injuries and were taken to the hospital.

Nobody was unaccounted for, but a dive team was doing a final sweep, just in case.
...
Photographs showed a wide gap in the northern side of the bridge, with significant debris in the water.

The bridge, built in 1955, has a sufficiency rating of 57.4 out of 100, according to federal records. That is well below the statewide average rating of 80, according to an Associated Press analysis of federal data, but 759 bridges in the state have a lower sufficiency score.

The bridge is classified as “fracture critical,” meaning it lacks redundancy and one big structural failure can bring the whole thing down, according to the Federal National Bridge Inventory.
 
Was the truck grossly oversized, or was the margin of safety razor-thin?
I'm sure a number of lawyers will debate that question soon enough.
The truck was not oversized by weight. It was oversized by height. The truck apparently hit the overhead horizontal bracing between the two main trusses (this is a through-truss span - where the supporting trusswork is above the roadway). If so, that resulted in the loss of stability in the main trusses and caused the failure.
 
Last edited by a moderator:
Here is a photo of the bridge (in better days). This is what is called a "through-truss" bridge. Instead of the main support structure being under the roadway, in a through design, it is above the roadway.

The truck is said to have been overheight and hit the steel that is located over the roadway. That steel keeps the two main support trusses in the proper position, and provides lateral wind load resistance. The main support trusses are at each side of the roadway. If the support trusses are not supported adequately, they lose stability.

i-005_nb_exit_229_02.jpg


Photo: AARoads
 
Last edited by a moderator:
It seems to me that the bridge should have been designed so that in the case of a strike, the structure would be resistant enough so as not to buckle upon impact.

Additionally, the principles of redunancy would suggest that if a beam was taken out by a vehicle strike (shouldn't happen, as above), the loss of that beam would not immediately result in structural instability.

Furthermore, clearances should be high enough to avoid most overheight vehicles. Interstate Highway standards indicate that the minimum clearance above the surface of the pavement is 16'. 14' is allowed in urban areas (generally to accommodate older infrastructure, such as this bridge, built in 1955). Absent 16' clearance, signage and other warning devices should be in place to warn drivers.

A check of streetview shows that no height warnings are in place prior to the exit (#229) north of the span. Additionally, no height warnings are in place on the bridge superstructure itself, which indicates that it probably does meet the minimum clearance of 16'.

Additionally, the overheight truck crossed almost the entire bridge before the span collapsed. The truck was traveling southbound. The span that fell was the southernmost span. The bridge has 4 identical through-truss spans. That means this truck crossed 3 of the spans without creating any problems. What was different about the fourth span?

(Conjecture):

I suppose it's possible that the truck changed lanes (the span curves downward toward the right shoulder and is highest above the center jersey barrier). But absent that, could the span have been sagging? Was there new asphalt on the last span that raised the level of the pavement. We won't know the cause for a while. But I don't think this is as simple as it first appears.
 
If the bridge collapsed because the top of a truck or the load in the back of the truck hit a single overhead span, that is not a robust design. Should not have a single point of failure like that.

Searching the news, Washington State is one of the states that has been debating a gas tax increase to pay for transportation and infrastructure projects. The state Senate was debating over a 10 cent gas excise tax increase.One part of the debate has been over that only 10% of the new gas tax revenue was slated to go to maintenance with the rest of it to new road and bridge projects. My bet is that the state legislature is now going to pass the gas tax increase with more going to maintenance. Replacing or modifying a bridge with a poor design by modern standards can fall under maintenance.

The I-5 bridge collapse could have some effect on the debate at the national level over increased federal funding for infrastructure which Obama keeps asking for and the House Republicans keep ignoring. But with the DC political scene consumed by grade B or C scandals of the week, the bridge collapse will likely be mostly forgotten by next week.

The bridge collapse took out the major road between Seattle and Vancouver BC. The Cascades trains could see a big boost in Vancouver travel in the next few months while the bridge is getting repaired. With 2 new Talgo train sets being delivered, how quickly could WA and Amtrak add an additional daily Seattle to Vancouver BC train? Normally, that would take a long time, but in an emergency, they may be able to do something quickly.
 
I-5 Skagit River bridge collapse caused by oversize load
MOUNT VERNON, Wash. (AP) - A truck hauling a too-tall load hit an overhead girder of a bridge on the major thoroughfare between Seattle and Canada, sending a section of the span and two vehicles into the Skagit River below, though all three occupants suffered only minor injuries.

It happened about 7 p.m. Thursday on the north part of the four-lane Interstate 5 bridge near Mount Vernon, about 60 miles north of Seattle, and disrupted travel in both directions.

Initially, it wasn't clear if the bridge just gave way on its own. But at an overnight news conference, Washington State Patrol Chief John Batiste blamed it on a tractor-trailer carrying an oversize load that hit an upper part of the span. The vertical clearance from the roadway to the beam is 14.6 feet.

"For reasons unknown at this point in time, the semi struck the overhead of the bridge causing the collapse," Batiste said.

The truck made it off the bridge and the driver remained at the scene and cooperated with investigators. Authorities have not yet said what the truck was carrying.

Two other vehicles went into the water about 25 feet below as the structure crumbled. Three people were rescued and were recovering Friday.
...
Traffic along the heavily travelled route could be affected for some time. The bridge is used by an average of 71,000 vehicles a day.
...
The bridge was inspected twice last year and repairs were made, Transportation Secretary Lynn Peterson said.

"It's an older bridge that needs a lot of work just like a good number of bridges around the state," she said.

Transportation officials are working on plans for either a temporary or permanent replacement, she said.

The National Transportation Safety Board was sending an investigative team.
...
The bridge was not classified as structurally deficient, but a Federal Highway Administration database listed it as being "functionally obsolete" - a category meaning that the design is outdated, such as having narrow shoulders and low clearance underneath.

The bridge, which was inspected last August and November, was built in in 1955 and had a sufficiency rating of 47 out of 100 at its November 2012 inspection, Transportation Department spokesman Noel Brady said Friday. The state average is 80, according to an Associated Press analysis.
...
The bridge was 1,112 feet long and 180 feet wide, with two lanes in each direction, Brady said. There are four spans, or sections, over the water supported by piers. The span on the north side is the one that collapsed. It's a steel truss bridge, meaning it has a boxy steel frame.
...
 
Was it one damaged truss member, or multiple? Did the bracing members just buckle, or did they tear-out at the connections? Did the failure buckle or tear the top chord of the truss in one or more places? Did other things happen that we haven't considered? Less than 24 hours after the failure, we don't know exactly what happened, so saying that it was a design problem is at the very least premature.

This bridge was built in 1955. In the 58 years since then, about 1 billion vehicles have crossed safely. A detailed structural assessment was made by WDOT in 2008, and found no problems. In the most recent American Society of Civil Engineers infrastructure report (released just a few weeks ago), this bridge was not among those found structurally deficient. In short, there is nothing in the design, the current reported condition of the bridge, or the reported traffic volumes that would identify a need for replacement.

Redundancy is a great concept (and plays well to the public), but building redundancy into a structure is expensive, and full redundancy is not practically achievable. There is always some level of damage, even single element failure, that can cause any structure to fail. How much should be spent to prevent a highly improbable failure? With this event, it is not just this through-truss bridge. It is every bridge like this: hundreds of bridges? Thousands of bridges? Replace or modify them all because of a once in a billion failure at one bridge?

By the way, bridge replacement or significant structural modification is never a "maintenance" cost. It is always capital.
 
State pushes for more Amtrak train service in wake of I-5 bridge collapse
The Washington State Department of Transportation is trying to add additional Amtrak Cascades train service to the route between Seattle and Vancouver, British Columbia, to help travelers circumvent the collapsed Interstate 5 bridge at Mount Vernon.
Currently Amtrak Cascades operates four trains daily on the route — two northbound and two southbound — and a goal is to double that if possible, said Ron Pate, director of the Transportation Department's Rail Division.
...
Each of the current trains carries 250 people, Pate said, and passenger cars aren’t available to lengthen those without taking capacity from somewhere else in the system. So the state is negotiating with BNSF Railway, and Amtrak, so see if more rail cars can be borrowed from somewhere.
It’s not clear how much adding the service would cost, or who would pay for it, Pate said, although the state does have an emergency management fund.
While the state would like to add the capacity for Memorial Day weekend, it wasn’t clear Friday morning if this would be possible, he said.
 
It seems to me that the bridge should have been designed so that in the case of a strike, the structure would be resistant enough so as not to buckle upon impact.
Maybe we should also redesign our trains so they can plow into improperly driven commercial vehicles without being damaged or derailing?
 
Last edited by a moderator:
State pushes for more Amtrak train service in wake of I-5 bridge collapse
The Washington State Department of Transportation is trying to add additional Amtrak Cascades train service to the route between Seattle and Vancouver, British Columbia, to help travelers circumvent the collapsed Interstate 5 bridge at Mount Vernon.
Currently Amtrak Cascades operates four trains daily on the route — two northbound and two southbound — and a goal is to double that if possible, said Ron Pate, director of the Transportation Department's Rail Division.
...
Each of the current trains carries 250 people, Pate said, and passenger cars aren’t available to lengthen those without taking capacity from somewhere else in the system. So the state is negotiating with BNSF Railway, and Amtrak, so see if more rail cars can be borrowed from somewhere.
It’s not clear how much adding the service would cost, or who would pay for it, Pate said, although the state does have an emergency management fund.
While the state would like to add the capacity for Memorial Day weekend, it wasn’t clear Friday morning if this would be possible, he said.
Not to take away from more trains, but how did people drive between Seattle and Vancouver before there was an Interstate? Surely there are some state highways that could be used.

The small town I live in is bisected by a US highway that parallels the Interstate for almost a 1000 miles. We sometimes prefer to drive on it rather than the Interstate as it is less crowded and safer. Whenever there is a bad accident on the Interstate the police just detour everyone over the US highway. People in town love it as it drums up more potential customers for all the businesses.
 
Not to take away from more trains, but how did people drive between Seattle and Vancouver before there was an Interstate? Surely there are some state highways that could be used.
US 99 was turned into I-5 starting in 1957. So there is no parallel US highway in the area. State Route (SR) 9 does parallel I-5 nearby, and there are some other highway options as well.
 
Not to take away from more trains, but how did people drive between Seattle and Vancouver before there was an Interstate? Surely there are some state highways that could be used.
US 99 was turned into I-5 starting in 1957. So there is no parallel US highway in the area. State Route (SR) 9 does parallel I-5 nearby, and there are some other highway options as well.
Thanks, Charlie - that makes sense. I know there are other places where the US highway went away as the Interstate was built "over it". Didn't realize this was the case there in Washington.
 
Was it one damaged truss member, or multiple? Did the bracing members just buckle, or did they tear-out at the connections? Did the failure buckle or tear the top chord of the truss in one or more places? Did other things happen that we haven't considered? Less than 24 hours after the failure, we don't know exactly what happened, so saying that it was a design problem is at the very least premature.
This bridge was built in 1955. In the 58 years since then, about 1 billion vehicles have crossed safely. A detailed structural assessment was made by WDOT in 2008, and found no problems. In the most recent American Society of Civil Engineers infrastructure report (released just a few weeks ago), this bridge was not among those found structurally deficient. In short, there is nothing in the design, the current reported condition of the bridge, or the reported traffic volumes that would identify a need for replacement.
It was, however, considered "functionally obsolete" because there was no redundancy build into the structure:

"This is what they call a fracture-critical bridge, which means if any part of the bridge fails, there is no redundancy built into it so it will all collapse," CBS News Transportation Safety Analyst and former NTSB Chair Mark Rosenker tells KIRO Radio Seattle's Morning News.
The Interstate 5 bridge that collapsed into the Skagit River was built in 1955 and was inspected two times last year in August and November.
Transportation Department spokesman Noel Brady in Seattle says it had a sufficiency rating of 47 out of 100. The state average is 80, according to an Associated Press analysis.
http://mynorthwest.com/11/2281457/Collapsed-I5-bridge-characterized-as-functionally-obsolete-fracture-critical
 
The Talgo trains that normally ply the route can't be lengthened. So once you factor in the number of people that would have ridden anyway,

the twice-daily Talgo service probably won't get much of a boost. And there are no additional Talgo sets in service w/o taking away a train from

somewhere else on the Cascades schedule.

That said, you could do one of several things. You could add an additional service using non-Talgo equipment. Or you could swap a Talgo set

for a single-level coach set. Both of those assumes such equipment is available. For utilization purposes, you wouldn't have to run the additional

train all the way up to VAC, you could turn it in Bellingham. Running an additional train also requires the host railroad to sign off on the plan.

There are definitely alternate highway routes -- multiple, in fact. The question is how many of the 70,000 daily vehicles that use the bridge will

attempt to use those routes. None of those routes is really set up to accommodate anywhere near the level of traffic carried on I-5.

Another Amtrak note: This bridge was used by the multiple-daily Thruway bus services that go up from SEA to Bellingham and Vancouver.
 
Another thing you could do, train-wise, is temporarily extend some Sounder trains up to Bellingham. Again, this requires host railroad

approval. But the equipment is certainly there. They could extend some of the evening outbound runs up to Bellingham.
 
So, what are the chances this becomes a permanent addition to the system? I ask that because it's likely to take the better part of a year, at the very least, to replace the bridge...and possibly longer.
 
So, what are the chances this becomes a permanent addition to the system? I ask that because it's likely to take the better part of a year, at the very least, to replace the bridge...and possibly longer.
I bet they can replace that span a lot quicker then that.

Assuming the adjacent spans were not damaged by either the truck or by the collapse and assuming the original shop detail drawings are available for the steel, getting a new span fabricated should not be that big a deal. Then it's a matter of demolishing and removing the wreckage and mobilizing equipment to install the new steel and pour the deck. The real time consuming aspect of typical bridge construction - the piers and foundations - are fine. I could be way off (wouldn't be the first time), but I put the over/under at 12 weeks assuming an in-kind replacement of the one collapsed span.

Now, if they decide to use the failure as an "opportunity" to put in an entirely new bridge (or, politically, they want don't want to replace in-kind), then a year is possible, but even that is at the outside. The I-35W bridge in Minneapolis, a much larger and more complex structure, was replaced in exactly one year less than 14 months from the collapse, and that used a completely new design for the superstructure.

Correction: it was one year from when the design/build contract was let. The contract was let seven weeks after the collapse.
 
Last edited by a moderator:
NTSB: Bridge collapse in Wash. is wake-up call

State transportation officials began working on both a temporary solution and a permanent fix within hours of the bridge collapse, he said.

The goal is to get I-5 open as quickly as possible, while making sure the solution is as safe as possible, he added.

Officials were looking for a temporary, pre-fabricated bridge to replace the 160-foot section that failed, Gov. Jay Inslee said Friday. That option could be in place in weeks. Otherwise, it could be months before a replacement can be built, the governor said.

Inslee said it will cost $15 million to repair the bridge. The federal government has promised $1 million in emergency dollars and more money could come later, according to Washington's congressional delegation.
 
Back
Top