I -40 Hernando Desoto Bridge at Memphis shutdown

[kalvado]

I still do not see where this member actually supports anything. There are far too few rivets actually holding it to the support (rod or cable). I still believe if it has any function at all it is  horizontal stabilization. Meaning keeping the actual deck supports from torsioning (twisting, dancing ...think swinging your hips doing the twist but on a MUCH smaller scale). I think this feature is primarily to support the deck support beam during a maintenance operation that removes the suspension cable / rod. It would also lend some redundancy if a support cable / rod failed. This torsioning would explain the break too.

This design is too far off the beaten path to just say, this is how the books say it works.  I realize I am only positing guesses.  I want to figure out how the design works maybe to confirm my theories, and maybe to see what the real risk actually was. I am not going to suggest that they should have left it open, but really knowing how serious it actually is / was.

As to the 2019 video, it was not what he was looking at/for. He may have fast forwarded to the part he was looking for. The cable and the truss.  Even if he had seen it as he was quickly looking for what he was tasked to look at, he might have mistaken it for a scratch in the paint, or debris, even bird poop (yes, I know that bird poop is a stretch.)

Again, I wish I knew how the weight is supposed to be distributed on this thing.

Thank you for the educated guesses - this is significantly more than I can come up with myself. If you come across any details, I - for one - would greatly appreciate sharing them.

I wonder if that beam was classified as "fracture critical" though. Bridge shutdown implies that.
If so, requirement is to have hands on inspection, when inspector gets at arm length (<3') of the inspected spot. So biannual close inspection - drones seem to be not suitable to the regulation - is a must. There is a rumor Arkansas specifically reduced that to a year (but maybe covid?)
Anyway, fired inspector purportedly missed the crack at least TWICE - and drone footage is just a proof that inspection really missed that and the crack was there during previous cycles. Drone was clearly further away and not zooming to "arm length" quality. 
Which brings up a question if being a  licensed professional engineer and certified bridge inspection team leader implies knowing this information...

UPD: some digging - Of course the bridge is classified as fracture critical, FC inspection date is September 2019, inspection interval 24 months, although AR indeed conducts FC inspections annually. Looks like 2020 inspection was not submitted to FHWA.  Structure number 79I00400001 for TN and 000000000005141 for AR.  Drone footage was taken in May 2019 and was explicitly not a part of FC inspection.

[The Ghostbuster]

I don't like how Interstate 40, on the east end of the bridge, drops from six lanes to four lanes through the Riverside Dr. interchange. It should be six lanes all the way through. Also, I have noticed that the eastbound off-ramp and the westbound on-ramp to and from Front Street have stub ramps: https://www.google.com/maps/@35.1529006,-90.0522629,473m/data=!3m1!1e3. Were those stub ramps to have gone to an unbuilt freeway in Memphis, such as an extension of TN 300 or an extension of Riverside Dr.?

[bwana39]

I don't like how Interstate 40, on the east end of the bridge, drops from six lanes to four lanes through the Riverside Dr. interchange. It should be six lanes all the way through. Also, I have noticed that the eastbound off-ramp and the westbound on-ramp to and from Front Street have stub ramps: https://www.google.com/maps/@35.1529006,-90.0522629,473m/data=!3m1!1e3. Were those stub ramps to have gone to an unbuilt freeway in Memphis, such as an extension of TN 300 or an extension of Riverside Dr.?

The HDB was originally 4 lanes (2X2) they took the shoulders and narrowed the lanes into 3x3.

[EngineerTM]

I still do not see where this member actually supports anything. There are far too few rivets actually holding it to the support (rod or cable). I still believe if it has any function at all it is  horizontal stabilization. Meaning keeping the actual deck supports from torsioning (twisting, dancing ...think swinging your hips doing the twist but on a MUCH smaller scale). I think this feature is primarily to support the deck support beam during a maintenance operation that removes the suspension cable / rod. It would also lend some redundancy if a support cable / rod failed. This torsioning would explain the break too.

The two main truss spans are a through arch/tied arch bridge.  The best way to think of how a tied arch bridge behaves is similar to an archery bow.  The upper truss framing is the compression side of the bridge (like the bow) and the bottom tie girders is the tension side of the bridge (like the bowstring).  Think about what happens if you pull on an archery bow and the bowstring suddenly breaks - the bow itself suddenly snaps itself out of its bowed shape.

This was the structural risk to this bridge.  If that lower tie girder was to have fractured completely, the upper truss framing would have lost that critical support and could have progressively and rapidly failed as one or more of those members buckled.  Photos released from TDOT showed that the bottom plate of the box was still intact, but the crack was starting to work itself through.

This is the simple way of looking at it.  Because this through arch/tied arch bridge does not have load path redundancy, a complete failure of the lower tie girder could have caused either a partial or full collapse of the bridge because the "bowstring" would not have been tying the entire structure together.

As to your question about the number of fasteners - steel is very strong in tension and you don't need a lot of material to make it work, structurally.  Where steel gets into trouble is in compression; the steel shapes tend to buckle, which is why there is so much additional frame bracing.

[EngineerTM]

As to the 2019 video, it was not what he was looking at/for. He may have fast forwarded to the part he was looking for. The cable and the truss.  Even if he had seen it as he was quickly looking for what he was tasked to look at, he might have mistaken it for a scratch in the paint, or debris, even bird poop (yes, I know that bird poop is a stretch.)

It is very unfortunate that the inspector either taking or reviewing the drone video did not fully appreciate the severity of what was revealed (whether he was rushed or otherwise).  The proper course of action would have been to pause and carefully look at those frames to see what that was.  Doing one's due diligence as an inspector should have been, at least, to make a couple phone calls and ask for a follow up field check to verify what that anomaly was.  I watched that video; it's not like it was a quick glancing flyby that was only in the frame for a couple seconds.  It was visible for longer than that and even to an untrained eye, not easy to miss.

Regardless, this does not excuse the negligence of whoever was responsible for the arm's length fracture critical inspection prior to this drone inspection.  If it is later determined to be a fatigue crack, then such a crack should have been detected and acted upon earlier when it likely would have been less costly (in terms of material and labor costs, as well as the high economic costs) to correct.

[cpzilliacus]

https://www.tn.gov/tdot/news/2021/5/17/tdot-awards-contract-on-the-i-40-hernando-desoto-bridge-repair.html

Kiewit Infrastructure Group selected to fix I-40 bridge.

There is a link in that press release that could be of interest:

Interstate 40 Hernando DeSoto Bridge:  https://www.tn.gov/tdot/projects/region-4/i-40-hernando-desoto-bridge.html

[cpzilliacus]

The two main truss spans are a through arch/tied arch bridge.  The best way to think of how a tied arch bridge behaves is similar to an archery bow.  The upper truss framing is the compression side of the bridge (like the bow) and the bottom tie girders is the tension side of the bridge (like the bowstring).  Think about what happens if you pull on an archery bow and the bowstring suddenly breaks - the bow itself suddenly snaps itself out of its bowed shape.

This was the structural risk to this bridge.  If that lower tie girder was to have fractured completely, the upper truss framing would have lost that critical support and could have progressively and rapidly failed as one or more of those members buckled.  Photos released from TDOT showed that the bottom plate of the box was still intact, but the crack was starting to work itself through.

This is the simple way of looking at it.  Because this through arch/tied arch bridge does not have load path redundancy, a complete failure of the lower tie girder could have caused either a partial or full collapse of the bridge because the "bowstring" would not have been tying the entire structure together.

As to your question about the number of fasteners - steel is very strong in tension and you don't need a lot of material to make it work, structurally.  Where steel gets into trouble is in compression; the steel shapes tend to buckle, which is why there is so much additional frame bracing.

Thank you for the thorough and lucid discussion of why the crack in that member is such a big deal.

Should fracture-critical bridges get an inspection every 12 months instead of every 24 months?  Especially  bridges that were constructed in the peak of Interstate construction in the late 1950's though about 1970 and have components that are fracture-critical - bridges that are now 50 to 60 years old and probably carrying heavier loads from laden trucks than those allowed when the the bridges were engineered.

[davewiecking]

https://twitter.com/memphonewslady/status/1394805035817709572
So the crack was forming in 2016....

[kalvado]

The two main truss spans are a through arch/tied arch bridge.  The best way to think of how a tied arch bridge behaves is similar to an archery bow.  The upper truss framing is the compression side of the bridge (like the bow) and the bottom tie girders is the tension side of the bridge (like the bowstring).  Think about what happens if you pull on an archery bow and the bowstring suddenly breaks - the bow itself suddenly snaps itself out of its bowed shape.

This was the structural risk to this bridge.  If that lower tie girder was to have fractured completely, the upper truss framing would have lost that critical support and could have progressively and rapidly failed as one or more of those members buckled.  Photos released from TDOT showed that the bottom plate of the box was still intact, but the crack was starting to work itself through.

This is the simple way of looking at it.  Because this through arch/tied arch bridge does not have load path redundancy, a complete failure of the lower tie girder could have caused either a partial or full collapse of the bridge because the "bowstring" would not have been tying the entire structure together.

As to your question about the number of fasteners - steel is very strong in tension and you don't need a lot of material to make it work, structurally.  Where steel gets into trouble is in compression; the steel shapes tend to buckle, which is why there is so much additional frame bracing.

Thank you for the thorough and lucid discussion of why the crack in that member is such a big deal.

Should fracture-critical bridges get an inspection every 12 months instead of every 24 months?  Especially  bridges that were constructed in the peak of Interstate construction in the late 1950's though about 1970 and have components that are fracture-critical - bridges that are now 50 to 60 years old and probably carrying heavier loads from laden trucks than those allowed when the the bridges were engineered.

This bridge was inspected on 12 month schedule, at least 2 last inspections - and maybe more! - didn't reveal the problem.
If anything, there should be a trend to better inspection methods as human eye is no longer the best available instrument. Strain gauges at critical points - a lot of them, fed into numeric models?. Eddy current and ultrasonic tools on a larger scale? AI analysys of inspection footage? Something else? Not sure.

As for older stuff... I found some report on Delaware river bridge, NJ-PA turnpike one, which had a similar crack few years ago. Fairly hair raising read I should say.

[kalvado]

https://twitter.com/memphonewslady/status/1394805035817709572
So the crack was forming in 2016....

Do I hear 2014? Cm'on, guys, we can find that!

[edwaleni]

There was a group who canoed up the Mississippi in 2016 and carried a camera to document for Google Maps.

If you follow their route, and zoom in on the bridge just before they go under, you can see a blurry outline of the fracture.

[ethanhopkin14]

There was a group who canoed up the Mississippi in 2016 and carried a camera to document for Google Maps.

If you follow their route, and zoom in on the bridge just before they go under, you can see a blurry outline of the fracture.

Link or it didn't happen.

[kalvado]

There was a group who canoed up the Mississippi in 2016 and carried a camera to document for Google Maps.

If you follow their route, and zoom in on the bridge just before they go under, you can see a blurry outline of the fracture.

Link or it didn't happen.

I guess this is more or less the image, give or take. I couldn't find the fault, though:
https://www.google.com/maps/@35.1525487,-90.0594502,2a,90y,341.8h,92.68t/data=!3m6!1e1!3m4!1sMJuyEhoFckHN_08QyU2CrA!2e0!7i13312!8i6656

[US71]

There was a group who canoed up the Mississippi in 2016 and carried a camera to document for Google Maps.

If you follow their route, and zoom in on the bridge just before they go under, you can see a blurry outline of the fracture.

Link or it didn't happen.

Que?

[codyg1985]

Finally, TDOT has updated one of the message boards for I-40 westbound traffic in Nashville to warn of the closure of the bridge.

[silverback1065]

I still do not see where this member actually supports anything. There are far too few rivets actually holding it to the support (rod or cable). I still believe if it has any function at all it is  horizontal stabilization. Meaning keeping the actual deck supports from torsioning (twisting, dancing ...think swinging your hips doing the twist but on a MUCH smaller scale). I think this feature is primarily to support the deck support beam during a maintenance operation that removes the suspension cable / rod. It would also lend some redundancy if a support cable / rod failed. This torsioning would explain the break too.

The two main truss spans are a through arch/tied arch bridge.  The best way to think of how a tied arch bridge behaves is similar to an archery bow.  The upper truss framing is the compression side of the bridge (like the bow) and the bottom tie girders is the tension side of the bridge (like the bowstring).  Think about what happens if you pull on an archery bow and the bowstring suddenly breaks - the bow itself suddenly snaps itself out of its bowed shape.

This was the structural risk to this bridge.  If that lower tie girder was to have fractured completely, the upper truss framing would have lost that critical support and could have progressively and rapidly failed as one or more of those members buckled.  Photos released from TDOT showed that the bottom plate of the box was still intact, but the crack was starting to work itself through.

This is the simple way of looking at it.  Because this through arch/tied arch bridge does not have load path redundancy, a complete failure of the lower tie girder could have caused either a partial or full collapse of the bridge because the "bowstring" would not have been tying the entire structure together.

As to your question about the number of fasteners - steel is very strong in tension and you don't need a lot of material to make it work, structurally.  Where steel gets into trouble is in compression; the steel shapes tend to buckle, which is why there is so much additional frame bracing.

Sounds like you're a structural engineer. thanks for the explanation! makes a lot of sense now! 

[kalvado]

Interestingly enough, there was an updated set of gauges, primarily for seismic monitoring, installed on a bridge in 2016. That includes a strain gauge on a cracked member.
I wonder if any anomalies would show up in that data, and if there would be useful for  future problem detection.

[cjw2001]

https://www.google.com/search?q=little+rock+ar+to+nashville+tn&ie=UTF-8&oe=UTF-8&hl=en-us&client=safari

I see Google reroutes EB travelers through city streets to connect back to I-40 once in Memphis.  I would think to stay on I-55 to I-240 East would be better.

Google directions are not static - they will vary depending on current traffic conditions.  It's likely that at the specific point in time you asked for directions this was the better choice.  Ask again a few hours later and you may get a different answer.

[US71]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

[kalvado]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

$1000 for 10 days of full-time training.
https://www.nhi.fhwa.dot.gov/course-search?course_no=130055

[bwana39]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

$1000 for 10 days of full-time training.
https://www.nhi.fhwa.dot.gov/course-search?course_no=130055

Many of NHI Web-Based Training courses are currently off line to address required updates. Our technical team is diligently working to complete the updates and the updated courses will be posted as they become available. The high access/high priority courses are scheduled to become available by early Summer of 2021. Expect all other courses by late Summer of 2021. We will provide announcements as content becomes available. We apologize for any inconvenience this causes and we appreciate your selecting National Highway Institute for your training needs

[US71]

https://www.wmcactionnews5.com/2021/05/20/stupp-bridge-co-sheds-light-efforts-repair-i-bridge/

Stupp Bridge Co. sheds light on efforts to repair I-40 bridge

[Big John]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

$1000 for 10 days of full-time training.
https://www.nhi.fhwa.dot.gov/course-search?course_no=130055

My employer paid more tan that for my training 22 years ago.

[bwana39]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

$1000 for 10 days of full-time training.
https://www.nhi.fhwa.dot.gov/course-search?course_no=130055

My employer paid more tan that for my training 22 years ago.

Actually they will do it on your site for as low as $790. per student. (I am not sure how many students the minimum is.....)

[kalvado]

Looking at documents that have been posted on-line leads me to think there isn't a lot of training when it comes to inspecting.

$1000 for 10 days of full-time training.
https://www.nhi.fhwa.dot.gov/course-search?course_no=130055

My employer paid more tan that for my training 22 years ago.

Actually they will do it on your site for as low as $790. per student. (I am not sure how many students the minimum is.....)

Minimum is 20, max is 30 per the fhwa page above.  Which is probably a mere cost of doing business.
If agency or company needs those certified inspectors and pais for the class, they should cover the compensation for students over those 2 weeks as well. That alone can easily exceed $1000 per student ($500/week = $26k annually).
Life in general is pretty expensive - but is still popular despite all costs.