Bridge Collapse in MN Part 2 8

[JAE]

The other thread was getting quite long so I thought I'd start a new one.

Here's a snippet from a news report today about some of the engineering thinking about dealing with the bridge in the months preceeding the collapse:

[blue]Documents obtained by the Star Tribune of Minneapolis for a story published Sunday reveal details of how officials decided to conduct periodic inspections of the bridge rather than repair it in the months before it crumbled.

According to the internal state Department of Transportation documents, officials were ready Dec. 6 to go ahead with a plan to install steel plates at several areas on the bridge as a patchwork fix amid reports that it was structurally deficient, as recommended by an outside consulting firm. The project was shelved after the state determined the process could actually weaken the bridge.

Instead, officials decided in January to go with periodic safety inspections that would look for any cracks in the beams that would warrant emergency repair. Senior engineer Gary Peterson said contractor URS Inc. assured them that any cracks could be detected before they posed a serious safety risk.

Inspections of 52 steel beams began in May but were suspended when concrete repairs began earlier this summer.

The inspection strategy was also deemed to be more cost effective, but Peterson and state bridge engineer Dan Dorgan denied that money played a role.

Engineers were to have met Aug. 20 to discuss whether the inspections were effective or if they had to go back to the plating idea.[/blue]

 

[miecz]

Have you read the fine print on the software used to design our structures? I can tell you from experience, that some bridge software companies limit their liability to essentially - zero.

We designed an interstate bridge using a well known software package, and got burned when the software failed to check a critical code provision. After much legal maneuvering, we had to pick up the cost of reinforcing the bridge.

 

[prost]

Sticking up for the software people, as much as I sometimes hate them, how can the software people take any liability? They can't control the bumbling of the user. We as engineers are still completely responsible (as we should be, IMO) for making sure the answers out the computer make sense, or that the answer takes the code into account.

 

[dik]

d2y: Stressful?... not to an existentialist <G>

Dik

 

[csd72]

Prost,

If the software doesnt do what it claims to do then there should be some penalty to the software company. Otherwise what is the motiivation for them to improve their software?

As engineers we shoud always check the computer output though.

csd

 

[miecz]

prost-

In our case, the bumbling was by the software writer. They admitted the error. The underdesign was small enough that the design seemed reasonable. There was no way to pick up the error without a detailed hand calculation. (Even then, the code provision was so subtle that the first hand calc missed it.)

I agree, for the reasons you state, that the software vendors cannot take any liability. This is why I no longer design with software, or spreadsheets written by others. There is simply no way to verify correctness in each case without a hand calc. I write my own MathCAD sheets.

 

[RARWOOD]

I like JStephen's idea about ever one staying on their side of the Mississippi River. Unfortunately I live in the Minneapolis area and cross the river on a daily basis.

Just on Tuesday of this week I crossed the river four times on three different bridges. Here it is pretty hard to go almost anywhere with out crossing over the river several times on different bridges.

 

[BSE05]

During my stint as a combat engineer we were taught how to classify bridges and post load ratings etc.

One classification and sign posting was called "risk crossing".

That would be an interesting sign to come across in your travels. Yesterday I crossed a box girder bridge in Fall River , MA. Box girders, rivets, and lots of rust, 100's of feet in the air.

I felt like it might of been a risk crossing. Nothing like a rusty rivet.

 

[BSE05]

I decided to look up this bridge in the NBI. It starts with a big red tag "This bridge is structurally deficient"

Built in 1965 with rivets, I didn't think they used them that late. They were gone when I came into the profession in 1970's.

Functional Classification: Urban Minor Arterial
Service On Bridge: Highway
Service Under Bridge: Highway-waterway
Lanes On Structure: 6
Lanes Under Structure: 2


Structure Length: 1761.7 m
Bridge Roadway Width: 28 m
Operating Rating: 44.1 Metric Tons
Navigation Vertical Clearance: 41.1 m
Minimum Vertical Underclearance: 4.42 m over Highway
Number of Spans in Main Unit: 3 Spans
Material Design: Steel
Design Construction: Truss - Thru


Number of Approach Spans: 23
Approach Material Design: Steel
Approach Design Construction: Girder and Floorbeam System


Deck Condition: Good Condition
Superstructure Condition: Poor Condition
Substructure Condition: Satisfactory Condition
Scour: Foundations determined to be stable for calculated scour conditions
Bridge Railing: Meets currently acceptable standards.


Structural Evaluation: Meets minimum tolerable limits to be left in place as is
Water Adequacy Evaluation: Equal to present desirable criteria
Estimated Total Improvement Project Cost: $145780000
Year of Project Cost Estimate: 2006


Average Daily Traffic: 89000
Year of Average Daily Traffic: 2005
Sufficiency Rating: 47 %

47% ?? what does that mean? Its probably 500 or more feet above the river---- along way down.

 

[Qshake]

SteelyLee, as you probably can surmise a sufficieny rating is a culmination of multiple factors some of which are structural, other geometrical and yet others are needs and purpose. However you look at it, closer to 100 is better than closer to 0.

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.

 

[bridgebuster]

A highway sufficiency rating is similar to a bridge rating; it works on a scale of 0 to 100%; 50% is considered "tolerable."

The rating is based in part on type of road - urban or rural; pavement condition; and traffic volumes.

 

[hokie66]

Giving a bridge an overall rating based on a weighted average of the components is silly. A bridge with a perfect deck and a badly corroded superstructure should be give a rating based on the superstructure.

 

[cvg]

hokie66 hit it on the head. a chain is only as strong as it's weakest link. same for bridges...

 

[ozbridge]

Having spent most of my working life with steel railway bridges, I have seen a number of "ratings" of the types above and all seem to be trying to link a risk of failure to a cost of restoring the bridge to a near-new condition - it can't be done with consistency. The reality is that SOME well maintained bridges are operating at high stress levels or develop serious local faults (eg active cracks in stringers) whilst SOME bridges with poor paint, rust and minor cracks are safer due to more structural redundancy and generally lower stresses.

 

[jike]

I wholeheartedly agree with hokie66. Structures usually fail because of one poor or deteriorated detail or a series of events but not because of their "overall condition."

 

[miecz]

I agree with hokie66. It's hard to believe rehabilitation decisions are made based on a weighted average.

SteelyLee

Could you explain how to look up a bridge in the NBI?