Minnesota Bridge Findings 4

[jike]

http://www.msnbc.msn.com/id/22663216/

 

[jike]

Gusset Plates found to be about half of the thickness that they should have been.

 

[DaveAtkins]

If you add load to a bridge, wouldn't you check everything, INCLUDING the gusset plates? But connections typically have a larger factor of safety than the truss members themselves, so perhaps that is why they weren't checked.

DaveAtkins

 

[frv]

The article was saying that they were under designed from the beginning (even ignoring the added load due to the renovations).

 

[msquared48]

Quoting a portion of the article here...

"Transportation Secretary Mary Peters was expected to issue an advisory urging states to check the gusset plates when modifications are made to a bridge — such as changes to the weight of the bridge or adding a guardrail, said a federal official with knowledge of the plans. The official spoke on condition of anonymity because Peters had not yet made the announcement."

"Currently, such calculations are done for the entire bridge, but not down to the gusset plates, the official said."

So, it seems to be implied that modifications to the original design were made, but the gusset plates not rechecked. For them to be undersized by 50% seems to be a pretty substantial "modification". And the statement that it is still the practice not to recheck the plates really concerns me.

Mike McCann
McCann Engineering

 

[jike]

The following is a photo of those "thin" gusset plates.

http://minnesota.publicradio.org/display/web/2007/08/09/gussetfolo/

It looks much thinner than anything that I have ever designed. They say these were only 1/2" thick!!!!

 

[HgTX]

"And the statement that it is still the practice not to recheck the plates really concerns me."

Me too. It is NOT standard engineering practice to ignore connections. Then again, it's a statement coming from some "official" speaking anonymously from stolen tidbits.

Hg

Eng-Tips policies: faq731-376

 

[Qshake]

In the bridge industry, especially at DOTs where knowledge and money are limited for bridges of these types, it is not common practice to check connections.

This is not the standard that the rest of the industry is held to. Bridge Building engineers typically are expected to check everything from load path to members to connections.

It is true that much of the software in use, BARS, Virtis etc do not provide detailed results to the connections but that shouldn't prevent someone from checking them.

Speaking of checking....the undersized plates should have been caught in the original design.



Regards,
Qshake

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

 

[civilperson]

Should we short the stock shares of Jacobs Engineering?

 

[JAE]

jike,

Did you note that in your link (which was way back in August, 2007) that the NTSB was saying that the focus on gusset plates was "way overblown"?

 

[jike]

JAE:

Yeah, I noticed that too!

 

[abusementpark]

One would think that intuition alone would prevent someone from using plates that thin.

 

[msquared48]

So much for NTSB reports regarding Aircraft incidents...

Or anything else for that matter.

Mike McCann
McCann Engineering

 

[DRC1]

This really does not make sense. The bridge worked for 40 years. The construction loads to support the decking operation sould not have exceeded the deck capacity. There must have been a trigger. Corrosion or a frozen bearing must have started this. Once it fell, the plates would look like spagehti no matter what caused it. It sounds like the Feds were hell bent on proving their boss's theory. I'm not buying it yet.

 

[Qshake]

DRC1 - As I read the report, they're not saying the plates alone are the reason for the collapse, just that the undersize plates represent a design flaw that contributed to the collapse. In the news conference the spokesman noted that they continue to look for the 'straw that broke the camel's back'.

Regards,
Qshake

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

 

[msucog]

http://www.asce.org/asce.cfm

http://www.ntsb.gov/Recs/letters/2008/H08_1.pdf

near the top of the asce website is a link and information to the ntsb findings and recommendations

 

[bridgeart]

The "Adequacy of the U10 & L11 Gusset Plate Designs for the Minnesota Bridge No. 9340 (I-35W over the Mississippi River" report available from the link below provides detailed analysis of the main truss gusset plates:

http://www.ntsb.gov/Recs/letters/2008/H08_1_Design_Adequacy_Report.pdf

It shows in great detail calculations and assumptions used to obtain D/C ratios for the main truss gusset plates.

ok, Structural Wiki > Gusset Plate

 

[transmissiontowers]

I'm not a bridge guy so forgive my ignorance, but were there 1/2" gussets on both sides of the flanges? I can see the outer gusset plates but I wonder if they use another 2 on the inside of the flange that straddle the web.

When I design a butt splice in a tower leg member, I put an angle on the inside and 2 gusset plates on the outside legs so the bolts are in double shear. The total thickness of the inside splice angle and the outside gusset plates add up to more than the leg thickness so bearing is not a problem.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[hokie66]

When you look at the elevation of the truss in the report, it looks like there was a drafting error. The gusset plates in the main span may have been transposed with those in the backspan at the two nodes in question.

 

[JLNJ]

I'm with Hokie in thinking it was a drafting error.

Using a factor of safety similar to the other connection plates, those few plates which were 1/2" were probably supposed to be 1 1/2"!