I-35W Final Report due shortly 3

[bridgebuster]

NTSB Advisory
National Transportation Safety Board
Washington, DC 20594
November 6, 2008

SAFETY BOARD TO MEET, ADOPT A FINAL REPORT ON I-35W BRIDGE ACCIDENT


--------------------------------------------------------------------------------

The National Transportation Safety Board will hold a public Board meeting on its investigation of the I-35W bridge accident in Minneapolis, Minnesota on Thursday, November 13 at 9:30 a.m., in its Board Room and Conference Center, 429 L'Enfant Plaza, S.W., Washington, D.C. The meeting is expected to last two days.

About 6:05 p.m. central daylight time on Wednesday, August 1, 2007, the eight-lane, 1,907-foot-long I-35W highway bridge over the Mississippi River in Minneapolis, Minnesota, experienced a catastrophic failure in the main span of the deck truss. As a result, 1,000 feet of the deck truss collapsed, with about 456 feet of the main span falling 108 feet into the 15-foot-deep river. A total of 111 vehicles were on the portion of the bridge that collapsed. Of these, 17 were recovered from the water. As a result of the bridge collapse, 13 people died, and 145 people were injured.

A live and archived webcast of the proceedings will be available on the Board's website at

http://www.ntsb.gov/Events/Boardmeeting.htm.

Technical support details are available under "Board Meetings." To report any problems, please call 703-993-3100 and ask for Webcast Technical Support.

A summary of the Board's final report, which will include its findings, probable cause and safety recommendations, will appear on the website shortly after the conclusion of the meeting. The entire report will appear on the website several weeks later.

Directions to the NTSB Board Room: Front door located on Lower 10th Street, directly below L'Enfant Plaza. From Metro, exit L'Enfant Plaza station at 9th and D Streets escalator, walk through shopping mall, at the CVS store (on the left), and take escalator (on the right) down one level. The Board room will be to your left.

###

http://www.ntsb.gov

 

[hokie66]

Interesting that a replacement bridge has been designed and built before the report is complete on the old one.

 

[concretemasonry]

The report will be very interesting to see if they can find find a single reason or a combination consisting of original design, construction inspection, modifications and new loads, restoration precesses, unbalanced loads from construction in progress and general deterioration.

I am looking forward to the report since I was involved in the original bridge, drove over it for many years and fished under it (inspected and observed frequently for years).

The new bridge administrative/design/construction concept was a "no-brainer" because of economic factors. The design build with a early completion bonus was much more economical than the design, low bid and then construct condidering the quick need and the seasonal construction period.

It was good to see economics, design and construction used to gether. Now I know why economics was included as a required class for engineers when I went to school.

The upcoming report will have little immediate economic effect, but it could have some good long tern effects on design, inspection and maintenance, so that is why it had a slower time frame. (Not to defend the parties, but to understand the benefits)

Dick

 

[hokie66]

Dick,

Yes, I guess I was commenting more about the quick production of a new bridge than the duration of the investigation of the old. I also look forward to reading the report. Often more is learned from failures than from research. I have been one of many who have been guilty of speculation about the primary cause, that being in my mind simply drafting error in the original documents, but now we will know.

 

[concretemasonry]

The production of the new bridge was certainlt done fast and the combination of design and construction was very efficient. Everytime I had a chance to look at the 24x7 production od the segmnents and erection logistics, it said a lot for the abilities of the designers and contractor. They certainly spent a lot of overtime to take advantage of the predictable weather conditions.

I am not surprised at the time for report considering the beauracracy and politics of a public failure. The number of factors and history make it difficult to apply any real numbers to assess a single factor which the news media wants so they can draw out the "what ifs" of the other factors using expert consulatants.

Dick

 

[Qshake]

A recent leak of information and the Minneapolis Star Tribune pointed to only the underdesigned gusset plate as the reason for the bridge's collapse.

I would be very surprised if this is the only reason it is collasped and if so, would feel an injustice to the engineering community.

Regards,
Qshake

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

 

[bridgebuster]

I agree with you 100% QShake.

I've seen broken gusset plates on a suspension bridge stiffening truss but the cause was a built-in design flaw that created excessive torsion.

I've seen plenty of old trusses with deteriorated gusset plates that were still carrying heavy loads and functioning well.

I recall reading that some of the gusset plates were field welded to members as part of a repair program. If that's true, fatigue failure is a possibility.

Right now based on the published information I'm not convinced. I still suspect that the frozen roller nests had something to do with it.

 

[concretemasonry]

Only about 3 or 4 more days defore the report is released.

The prliminary reports I have read and heard say the "primary" cause was the gusset plates. This indicates there could be other contributing causes or combinations of contributing causes.

The under designed plates were an obvious factor that could not be eliminated as a factor. Because of the modifications, weathering and construction are other obvious items leading to the immediate collapse.

I hope the report includes the other possibilities and addresses the possible connections (no pun intended) to the collapse. We could all learn from the unfortunate tragedy/problem.

Dick

 

[DRC1]

I am sure that it was this sole gusset plate that caused the bridge to have a horrible rating, have the engineer advocate replacement ASAP (which Minn. DOT took to mean 12-14 years) need need anual inspections.
I have not read all the reports, and have not read very much at all about it lately, but although the gusset plate was deficent, was it ever determined that the plate was the cause of the collapse? Or is that what we find out this week?

 

[hokie66]

Yes, that is what we are waiting for. The undersized gusset plate issue did not affect the bridge rating, as it was an unknown fault, even apparently to the engineers who had previously done extensive reports on the bridge.

 

[bridgebuster]

The first report from the hearing:

http://news.yahoo.com/s/ap/20081113/ap_on_go_ca_st_pe/bridge_collapse_ntsb

 

[unclesyd]

Here is the webcast portal for the hearings if anyone is interested. The report will be available in two to three weeks.

The AP reported today that another factor that is being considered is the storage of 575,000 lbs of construction material in the area of one of the failed gusset plates. The AP source also stated that even with additional weight if the bridge had been designed right this additional weight wouldn't have been a problem. The contractor had permission to put the materials on the bridge from the MDOT.

http://www.capitolconnection.gmu.edu/ntsb/live/ntsblive-cc.htm

 

[unclesyd]

Some information today states the investigation is focusing on the upper gusset plates, especially at Node U-10. The biggest problem is trying to ascertain which of the plates failed first. Which I would assume to determine is the additional weight might have been a factor.

 

[unclesyd]

Correction to my last sentence in the my last post. I was talking on the phone and trying to digest, think and type
at the same time. I maybe able to do two but never three things at the same time.

Correction:
"Which I would assume is to determine if the additional weight on the bridge contributed to the failure.

From information I received they have practically ruled out problems with the bearings.

 

[concretemasonry]

unclesyd -

I agree that the major cause of the failures was mainly due to the proverbial "underdesigned gusset plates". There was no major corrosion and the bending on the plate was due to either initial construction or excess loads during the life and modifications, which will be hard to determine without the original design calculations. - Hindsite is always 20/20.

The bridge functioned well enough (although maybe not as designed) for 40+ years and was modified with some widening and new crash rails etc., through the years. This could have provided a false sense of security. In addition, the deck reconstruction, current construction activities and materials storage may have placed additional and unpredictable loading patterns since the traffic was also switched during the process.

I have not had a chance to read the report, so I can only offer some engineering opinions.

I.
From what I understand, the contractors were given "carte blanche" permission to operate on the deck during the reconstruction. It would be interesting to see if there was any real analysis on the possible effects on the old structure. The permission is understandable because of the variability of materials and mobile equipment. Since the traffic was diverted frequently and the construction process is variable, this could lead to some very interesting unbalanced loading conditions on an old structure. Since this was a contracted maintenance/upgrade, I assume the contractor had no knowledge of the actual structural condition and bid on the contract as presented. This may be behind the choice of a design/build contract for the new bridge.

II.
I have not been able to find out if any analysis of the structural capacity of the old bridge had been determined based on the recent condition.

III.
The "thin" gusset plate must have been due to the engineering calculations or a drafting error. Certainly it was not a cost cutting thing because of the minimal material cost, but it may have matched with the other members to create a clean detail to design, detail and build. Unfortunately, the details of the design are not available, so the parties can only rely on the information known or provided. It is unfortunate that the over-designed members and connections attract loads and ruin the conservative assumptions made by designers.

IV.
The NTSB report wil not end the subject. There will still be the costly law suits against the parties including, the MNDOT (unless they have immunity), contractors, subcontractors, people traveling in the opposite direction that contribute to the non-symetrical loading and even the auto manufacturers that produce cars to sense upcoming problems (sick joke, but they claim to be able to parallel park cars without driver assistance).


I hope we are able to sift out enough facts to learn from the disaster and prevent future ones that may occur.

Dick

 

[DRC1]

You bring up some good questions. We will know more when the report is released. However, here are my thoughts on your questions.

I The stockpiling of materials is common in bridge construction and is generally not investigated. This is based on the assumption that dumped materials are not pushed up much on bridges and occupy about the same area they would have in the truck, thus not significantly altering lane load. Staging I think is looked at from a structural point of veiw, but not in great detail. After 40 years, loads on the bridge shift and cutting it apart may change stresses in unaticipated ways.

II This is what I didn't like. Apperently the standard practice for widning the bridge is to check a few main members and if they are sufficent, it is assumed the whole load path is sufficent. My understanding is that when they modified the bridge the load path was not checked except for a few members and that even after the collapse, the feds were okay with that.

III My understanding is that it was probably a drafting error in that the wrong gusset plate was detailed for that connection.

 

[unclesyd]

If I may ask, why wouldn't a bridge or in fact any equipment that has been labeled "Structurally Deficient" warrant a little more attention if it is going to operate out side it's normal environment?
Or is it the "Devil is in the details" in this case.

 

[Qshake]

Good points raised by all. I do agree with unclesyd who asks why a structure that is labled structurally deficient, doesn't warrant additional structural analysis before, during and after major reconstruction.

Some will say that the bridge's structurally deficient rating (happening in a span of a single year when it went from 7 to 4 wherein it stayed until it's collapse) wasn't really addressing the structural capacity. True, in a way, because the rating is Federal technical language that the bridge should be on a list programmed for replacement. Since that process takes about 10 years to complete with all the environmental processes etc, the sooner it's on the list the better.

Others will say that the bridge was labeled structurally deficient because it caused some engineers to lose sleep over the myriad of poor conditions the bridge was facing.

I tend to be in the latter category and feel this bridge should have received more attention structurally before more weight was added, more traffic allowed, and conditions to continue to deteriorate. Such a evaluation would have, in likelihood found the underdesigned plates.


Regards,
Qshake

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

 

[hokie66]

I think we all agree in principle, except for one thing. This bridge was investigated, as reported in very extensive and presumably expensive reports, but the deficient gusset plates were not identified. Hopefully an explanation for that failing will be forthcoming.

 

[graybeach]

I found conclusion #12 (in the synopsis) that "Current Federal and State design review procedures are inadequate to detect design errors in bridges" to be very interesting. The associated recommendation involves owners developing procedures to detect and correct bridge design errors before the design plans are made final.

Does this mean that owners have to check calculations that have already been checked and sealed by a consultant? Either that or have an independent design reviews done by an alternate consultant?

Working for an owner of bridges and other infrastructure, I get calculations to review all the time, but there is no set procedure that covers what I should do with them. I have even been admonished by management for digging into calculations that have been PE sealed.

Usually I will do a few spot checks. If I find a significant mistake, I will plow through the whole thing. Or if the calcs call for reducing the capacity of some of our equipment, I will thoroughly check to make sure the reduction is really needed. My procedure is not really scientific, and sometimes I only look to see if the calculations seem to be well organized and referenced and if each page is signed by a preparer and a checker.

I find major errors in sealed calculations way too often, so I am looking forward to seeing what procedures that the FHWA comes up with.