It's being reported that the structural span portion of the bridge was just erected this past Sunday and that there were workers on the bridge when it collapsed.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

Another story

Link

Looks like it used the roof structure as the top chord of the truss. I wonder if the webs broke free from the bottom chord?

I wonder if the bottom section failed in tension then the top section buckled. The length to depth ratio looks very high. Were the cable reinforcement properly tensioned and were there provision for connecting and tensioning the reinforcing cables between section assemblies to unitize the assembly?

This article shows a rendering of the completed, cable-supported bridge.

This may a case where the team thought that the existing deck structure was strong enough to self-span the roadway while the cables and tower were erected. I wonder if the warm Florida sun hit it, made it expand, and the temporary construction load + expansion broke loose a metal-diagonal-to-concrete-web-chord connection...

http://www.miamiherald.com/news/local/community/mi...

"We shape our buildings, thereafter they shape us." -WSC

Quote (Lilliput1)

The length to depth ratio looks very high.

It appears to be shallow... but, not unreasonably so.

Dik

The bottom section looks continuous. It would be in tension, so a failure would leave a gap. The top chord is in compression and can't separate and looks continuous. This leaves the shear members and I expect this is a failure of one of the connections, either via shear or by punching though one of the decks. It's also possible the upper chord buckled.

The horrific part is that the bridge fell on a large number of cars that must have been stopped for a traffic light. It seems that at least 4 and possibly as many as 10 vehicles were under the bridge.

One witness reports that a support failed sideways, but I think that must have been in reaction to a failure elsewhere.

The lack of symmetry and the irregular columns is unsettling. See http://www.miamiherald.com/news/local/community/mi... Obviously without knowing the reinforcement making a call about sufficiency is impossible, but I am unnerved by the second column on the campus side; the side that fell on the cars.

Edit:
I see that this was to be cable supported. That explains the irregular beams and columns. Someone didn't do a good job of analyzing the partial construction strength.

Quote (mjb315)

I wonder if the warm Florida sun hit it, made it expand,

I noticed the heat mirage 'waves' in the news footage... I guess we'll have to wait until some technical discussion comes out... our thoughts go out to the survivors and the families of those involved.

Dik

It looks like the slab was post-tensioned. I wonder if the anchorage slipped. If so the (presumably) ungrouted conduit may have become the tragedy that some here warned was possible. Time will tell. It's certainly a horrible situation.

It is actually pretty cool weather for south Florida today. Currently 73F.

Quote (Archie264)

may have become the tragedy that some here warned was possible

Were there some warnings or concerns expressed on this?

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

From this article: https://amp.local10.com/news/florida/miami-dade/fl...

Quote:

The bridge is also made of self-cleaning concrete. When exposed to sunlight, titanium dioxide in the concrete traps pollutants and turns them a bright white, the university said.

I suppose titanium dioxide in the concrete wouldn't perhaps have any affect on strength, bond, etc. but I don't know.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

I noticed the waterway to one side where the abutment could be on pile...

Wonder if there was lateral movement from an eccentric loading causing the abutment to shift and the bridge to lose support on that end.

Mike McCann, PE, SE (WA)

Or maybe the structure was compromised during the moving/lifting process?

We'll see.

Mike McCann, PE, SE (WA)

MJB315 post shows link to what the completed bridge would look like. Since the cable stays and main support column is not yet installed a temporary support at the midspan of the deck (also center of roadway) should have been provided.

>>>Were there some warnings or concerns expressed on this?<<<

JAE, sorry, I didn't mean to imply that there were warnings regarding this specific structure; indeed, in my case this is the first I've heard of it. Rather, I was referring to the various discussions of bonded vs. unbonded post-tensioning that have occurred over time on the structural forum. I probably should have withheld speculation, though.

Looks like it is a temporary construction condition. Cannot see any sign of the second span or the cable stays etc.

You do have to wonder about the scheme. If the truss was supposed to be able to span simply across the roadway without the cable supports, why were the cable supports necessary at all? In the completed structure, it looks like the truss was to be continuous, and cable stayed. So the temporary condition was critical. I know, more loading in the permanent condition, but not enough to override the construction loading.

I agree and until Rapt pointed it out I didn't realize it was ultimately supposed to be a 2-span condition. And to have allowed traffic to go underneath it in the interim...?!

It will be interesting to learn what happened.

FYI regarding weather this week.

https://www.wunderground.com/history/airport/KMIA/...=

Nothing unusual about the weather. I think that can be discounted as a factor.

Here's an interesting "FINAL" design picture:
http://www.miamiherald.com/news/local/education/ar...

Was this thing a concrete truss? Hard to see from the photos if there were embedded structural steel members. Concrete trusses are rare, and for good reasons.

Was that truss constructed from concrete?

Dik

ya beat me by 10 sec...

Dik

It is was a concrete truss from the pictures. Also sayas an engineering company performed a stress test in the morning?

https://www.google.com/amp/abcnews.go.com/amp/US/p...

An article from heavy.com...there is a time lapse of the bridge move. Hopefully, some video at the time of the collapse will turn up.

https://heavy.com/news/2018/03/florida-internation...

The collapse occurred in seconds....non-ductile.....implies perhaps a shear failure on the one side that hit the ground. As others have noted, we'll have to wait and see. The structure was designed by Figg, a reputable bridge design group. It was an attempt at "accelerated bridge construction", a concept that involves pre-constructing and setting in place. Although touted as "new", that process has been used many times, particularly with steel frames.

Cable supported bridge collapse because no cables. Duh.

What a horrible tragedy. It could it have been prevented.

I am not a civil engineer, so bear with me as restate what is being discussed here, and what I have seen in reports, photos, videos and Google street view.

Here is the artist's rendering showing the cable supported pedestrian bridge.


Here is the main bridge span resting on two supports on either side of the highway. Notice NO cable supports.


The collapsed bridge shows no evidence of the tower. No tower = no cables. So, the span crosses 8 or 9 lanes of traffic with no supporting cables.


So maybe someone had built a temporary support from median strip to the bridge span? I'm looking through the pictures and video and seeing no evidence of any temporary support structure.

How could they have opened the road to traffic when the bridge had no cables? It's a cable supported bridge, for God's sake! And without any temporary support structure?

Am I missing something here?

Roopinder Tara
Director of Content
ENGINEERING.com

Report in the local news here that the structure was being "stress tested", whatever that means, at the time of failure. Don't know if that is true.

Mike McCann, PE, SE (WA)

hokie66 - That is the best of the various time lapse videos. They seem to spend an inordinate amount of time working on the bearing pads or something along those lines.

Here is a good shot of the post-tensioning Link

Notice the VSL Corp truck/van under the bridge Link
If VSL isn't working on this bridge, then they will for sure get to the bottom of what went wrong.

Notice also in the last photo the tensioning equipment. I wonder if they were tensioning some strand when something went wrong?

hokie66 - I have looked at several pictures and it does appear to be a concrete truss with post-tensioned bottom slab/chord. The truss web arrangement would appear strange, but looking at the final design, the webs match the angles of the permanent cable supports. Still they're not symmetrical at all for this temporary construction condition. It's a long span with certainly a lot of dead load and high truss web member forces. It's not a typical truss web arrangement, and it's a concrete truss. That's a lot of unconventional items in one large design.

I hate this situation for everyone--the injured, their families, the engineers, and our profession in general.

Hokie66 -- I don't know anything about this bridge specifically. In many cases, a pedestrian bridge may be adequate to self-support in a temporary construction condition, but a secondary support (like this cable stay) is desired for vibration and serviceability performance before it sees public traffic.

TLDR: Construction workers are a lot more tolerant of L/180 deflections than the public is.

Unfortunately, getting phased construction loading right is harder than it looks, so I agree that's a likely root cause. I'm glad to see the reputable names of Figg and VSL mentioned here. I think both those companies will be motivated to iron out what happened (rather than just rushing to rebuild).

----
The name is a long story -- just call me Lo.

RoopinderTara - I understand your frustration. This temporary span condition would have certainly been designed for... but either an error occurred in the design process or a construction error, or combination of the two.

It's a terrible tragedy that (as you have said) could have been avoided.

Colorado requires bridges in temporary conditions be inspected and signed off by a PE before public traffic is allowed underneath. Any idea if FL is the same?

If not, I bet it's coming!

----
The name is a long story -- just call me Lo.

The reports may be confusing stressing tendons with a stress test.

Quote (Ron)

It was an attempt at "accelerated bridge construction", a concept that involves pre-constructing and setting in place. Although touted as "new", that process has been used many times, particularly with steel frames.

It didn't appear to have any of the cable stayed structure in place. It may be that part of the procedure for erection was missing or missed; it failed 'way too soon', and it may have required temporary support that was 'missed'. In all my years... I've never done a concrete truss. I've seen a few concrete trusses used for old railway bridges, and, that's it.

If it was designed for cable stayed, then the members should be light to reflect the closeness of the stays. That feature may have provided a weak structure if not properly supported.

Dik

If I had to bet money on it at this point, I'd go with inadequate stability bracing of the top chord. What's the unbraced length of the top chord for buckling about the vertical axis? To make the unbraced length only the length of the bridge, the end columns would have to be stiff and strong to provide nodal braces, to use AISC's Appendix 6 term.

For what it's worth, someone apparently heard a loud "cracking whip" sound while stopped in his car under the bridge, several hours before the collapse: Link

Did the design engineers specify a phasing plan such or did their specification require the contractor to provide adequate temporary supports as required?

I'm guessing this has to do with construction phasing. Some lack of communication between the simple span vs continuous phases.

Reports are, that the PT tendons loosened and were being re-pulled. I am no expert on PT. Is that something that is common?

Also, video is out. The collapse was brittle.

https://www.google.com/amp/amp.miamiherald.com/new...

Video in link

[This 5 second video from a random cell phone in a building across the street captures the actual collapse of the structure]

All the end reaction went up that first diagonal in compression, and the next diagonal couldn't take it. Maybe a failure of the anchorage of that diagonal, or maybe incidental moment in the connection. A concrete truss was a poor choice. Many of these have been done in steel, which provides the ductility necessary to cope with things a bit out of alignment. I know that Figg has done many bridges, some a bit heroic, but I think they would be wishing they had not done this one.

From the video I think the first diagonal failed. There seems to be a wrinkle near the top just before interlace blurs between the slight deformation and the total collapse. This happens just before the video progress indicator touches the sidewalk in full screen.

I wonder if the reason so many people were standing just above the failure point was they had spotted a crack and were trying to figure that out - otherwise it's an even more horrible coincidence.

Edited - so they were apparently performing some work as noted by later posts.

At the time of collapse it appears they were stressing PT bar/s to the first COMPRESSION diagonal from the top flange, at the first interior 'node' from the future stay-cable tower, as per the following photos:

Future stay-cable tower to the left of the photo:




Photo looking along the longitudinal-axis of the bridge:




General view during erection:

After they set the bridge in place, they de-stressed the two top deck tendons. Link

The column at the end of the bridge where the tower was to be built, was only half of its final width. Once the other half of the bridge was in place the two half columns were to be joined together & the column would have had an octagonal shape.

Here is an interesting angle Link

Couple other large scale reference photos
Link

Link

Assuming there wasn’t a problem that was getting worse as more time passed, it would seem more likely for it to be related to what they were currently doing since it had been stable up until this point under similar gravity loads and weather.

So much for it lasting 100 years!! Is it true the firm was being sued for a platform failure back in 2016?

Why would they be stressing at that stage? The truss was spanning full length, and should have been complete, at least for that stage of construction.

Report mentioned above said that cables had loosened and were being retightened.

usually there is a combination of different 'things' when a collapse like that happens ...not just a single source....

I don't understand the detensioning of the top deck PT.

Most of interest would be the details of the anchorages within the concrete elements. Were they castings?

An issue with a concrete truss is that it doesn't want to act only like a truss, but also like a rigid frame, so there are substantial moments at the joints. Central stressing would not cope well with these moments, so a lot of deformed bars would be required. Does anyone other than me think that the photos show very little in the way of non-prestressed reinforcement?

Its a cable stayed bridge, that weighed 940 tons (if we can accept that the news reporting is correct).

If the bridge was designed for 100psf of traffic on it, that is an additional 340 tons approximately. So, dead load is by far the dominant load.

Without knowing project details, it seems that the simple answer is that the bridge probably wasn't designed for simple span in construction phase, so the engineer would have assumed shored construction until cables installed. The contractor didn't understand that.

Its possible also that the strand tensioning work at bottom chord was assumed to have already been finished, prior to releasing the shoring towers.

My kids brought back souvenir statuettes from Mexico, that appeared to have been made of clay.
... that had been extended with, er, locally sourced used food, if you catch my drift.
The statuettes had broken in shipment, in a suitcase, protected by dirty laundry.
Impromptu testing caused further fractures, simply by shaking the object gently while cantilevered from a hand's grip.
Those fractures looked a lot like the TV images of the broken bridge.

So as to infinitesimally reduce my colossal ignorance of related subjects, I ask:

Does Accelerated Bridge Construction normally include moving a concrete structure before the post-tensioning strands had been fully stressed and locked?

Is 'de-stressing' a normal procedure in PT construction, ever?

Is 'de-stressing' or 're-stressing' or 'stressing' a PT tendon such a trivial undertaking that it's often carried out on a structure with civilian traffic passing beneath?

Mike Halloran
Pembroke Pines, FL, USA

Is my recollection of statics right - that the diagonal that failed would have a compressive load that would roughly be half the weight of the bridge divided by the sine of the angle (assume 30 degrees, so 0.5) it makes with the horizon? If so then this puts a 940 ton load on that small (18 inch X 18 inch) section. And then it appears they added an additional compression load via the cable tensioner which appears to feed to the diagonal?

So that puts it at 6ksi for the dead load (24 inch x 24 inch drops it to just over 3ksi.) That seems to be in the middle to a bit above middle end of the capability of typical concrete.

If the load was eccentric that would make this even worse.

The loading regime of that 'truss' would be substantially different when loaded as a simple span or a cable stayed. There would likely be no similarity. As noted above, as a simple span, the end diagonal would be taking approximately half the span in compression. As a cable stayed structure, this could even be a tension member.

It will be interesting when the report on the collapse is completed, how many of the comments in this thread 'ring true'.

The bridge was under construction and it is not known what component was contractor's 'means and methods'.

Dik

Ingenuity...great photos...and as Hokie66 mentioned...why would they be stressing that compression diagonal in the field? It looks like there is a single tendon running thru the center of that diagonal...Why wouldn't there be reinforcing around the perimeter of the compression diagonal to increase it's moment capacity?

I don't know how many times I've had to hear about the Hyatt Regency Failure in getting my Ethics hours...but (and I am assuming here) If there were problems with slipping of tendons...and concrete cracking observed and they were trying to re-stress a tendon...then the Traffic should have been shut down....but then there is the other side...the Pride of our design and the Fear of letting others know we made a mistake...

I also wonder what happens to Figg as a company...there are post-tensioned bridges in our state done by them...otherwise a fine, outstanding company I'm sure...but they get work through State DOTs...Do state bridge engineers hire them as consultants...the public will already be concerned about any bridge done by Figg...With the lawsuits and damage of their image...If I was working for them...I'd by looking for a new employer....my $0.02

Quote (MikeHalloran)

Does Accelerated Bridge Construction normally include moving a concrete structure before the post-tensioning strands had been fully stressed and locked?

Is 'de-stressing' a normal procedure in PT construction, ever?

Is 'de-stressing' or 're-stressing' or 'stressing' a PT tendon such a trivial undertaking that it's often carried out on a structure with civilian traffic passing beneath?

Yes, No, No.

Yes.. it's normal to under stress the strands during the construction process and tensioning more after the instalation of beams for example.
No.. Haven't heard of de-stressing... re-stressing is a common method.
No.. these are dangerous activities and must be done without traffic.

Are we sure it was the first diagonal in compression that failed? What about the second diagonal which is running the opposite way and is in tension? How do you handle the connection at the bottom between the tension member and the bottom chord of the post tensioned slab? Is there a tendon or two that run thru a sleeve and then anchor into tendons running thru the diagonal?

It looks from the video like the bottom slab/chord buckled in flexure between the first and second diagonal.

Here is a link of the proposal from the design build team. I have a hard time imagining that the project would progress to construction without a design that considered the simple span condition before the the back span was placed and the two were tied together. I don't design bridges but aren't precast girders designed for simple span DL and continuous span LL (assuming 2+ spans)?

https://facilities.fiu.edu/projects/BT_904/MCM_FIG...

Well it's certainly a gorgeous design but I guess the devil was in the details, unfortunately. Complexity begets failure, to the sorrow of us all.

By those drawings, stressing a bar could cause eccentricity on the diagonal.

EngineerEIT...Great Find!!!!

On Sheet 115/173..the first compression strut is labeled #11...It looks like there is no P.T. strand for that member...but there has been talk of that member Tensioned in the field???

Member #10 shows 280k and is in tension...I see the connection detail at the top...but what about the bottom where it ties into the tendons running thru the deck?
Member #9 (in compression)and doesn't indicate any PT Tendons...Was there mild reinforcing in the compression member???

Funding was largely federal, from the previous administration:

Collapsed FIU Bridge Was Funded by Federal Grant Program Criticized for Shoddy, Politicized Review Process
The TIGER grant program has come under fire for putting politics ahead of technical concerns.

By Christian Britschgi
March 15, 2018

TIGER was created as an economic stimulus measure under President Barack Obama and morphed into a permanent program. It has awarded $5.6 billion in nine rounds of grants since 2009. Members of Florida's congressional delegation publicly lauded the TIGER award to FIU.

"Thanks to this TIGER funding, FIU students will be able to walk from their student housing to class through a pedestrian bridge across Southwest Eighth Street," Rep. Debbie Wasserman-Schultz (D-Fla.) said in 2013. "More jobs will be created in our community thanks to this grant, and I look forward to celebrating the project's success with everyone in South Florida."

Rep. Mario Diaz-Balart (R–Fla.) made similar comments on Saturday. "FIU has come a long way since the TIGER grant that funded this pedestrian bridge was awarded in 2013," he said. "This project represents a true collaboration among so many different partners at local, state, and federal levels, and in both the public and private sectors."

I would assume that the bridge under construction and transportation had intermediate supports and a corresponding pretension of the strands....when this section of the bridge was in place and the intermediate supports were removed then it experienced for the first time the forces and deflections caused by spanning the full length of the first span resulting in more compression in the top chord and thus a lessening in the tension present in the strands and thus the attempt to re-tension them...for a simple single span the top chord would always be in compression so why is the added tensioning required at this stage in construction...could it be possible that the temporary compression in the top chord due to the single span during construction plus the added design tension in the strands based on the completed bridge have contributed to the collapse...again, this is all speculation on my part, at the moment....

Quote (hokie66)

I don't understand the detensioning of the top deck PT.

I think the tendons being de-tensioned are TEMPORARY PT bars in the top of the top flange- the wheeled trailer support was centered about the first interior node point, so maybe that 'overhang' necessitated the temporary PT.

And this would explain the stressing operations occuring in the collapse photo - the photo above I posted may be the crew de-stressing the top bars (NOT the compression diagonal, as I first stated) - there is a large stressing stool in front of the ram - that is often used to access the nuts from threaded PT bar and therefore enable destressing.

For a short time I did inspections on elevated PT slabs and Ive seen them snap from over stressing during tensioning. It can be very violent and destructive to the concrete. Maybe a cause? I believe they said they were tensioning it when it collapsed right? One report said someone heard a loud "cracking whip" sound a few hours before and I know these things are loud when they snap. If its been up since Saturday what else would of caused this? Vibrations from the traffic? There must of been some type of external force that caused this right? I dont design bridges but if the design wasn't adequate enough to resist at least its own weight do you think they would of noticed cracking, deflections, ect. once they removed the supports?

I saw a traffic cam video of the collapse. They were tensioning tendons in the first diagonal when there was a sudden collapse, apparently buckling the top chord plate. Under a simple span condition, that would be a compression strut. Under the cable stayed condition, there might be load reversal (depending on the support scheme)....thus a need for tensioning. Any possibility they tensioned this member out of sequence? Tensioning of this member should probably have waited for the cable stays.

We're all guessing at the reasons....could be all or none, but certainly interesting though incredibly tragic. Unfortunate for all involved and their families for their loss. This will gone on for years to come.

From looking at drawing package erection procedure (page 125):
Link
* They were supposed to tension the strands in the longitudinal direction prior to setting main span and removing shoring.
* It was planned to set the main span, and 2nd span, and only then install the guys (which were 16" pipe).

Presumably they would have checked the engineering of the "main span as simple span", as its shown in their drawings and also in their construction schedule.

Its not clear to me whether they were re-tensioning the long strands or not. If they were, then this was either a sequencing screw-up (extremely unlikely in my opinion), or, when they removed the shoring they must have had structural problems and were trying to fix it by re-tensioning strands etc.

So based on the above, its possible that the construction sequence was followed, but the performance of the single span action was lacking, whether it was an engineering screw up or a materials screw up.

Is it possible that there were multiple tendons in web member and may have needed to be stressed in a balanced manner. Maybe that sequencing wasn't followed and all tendons on one side got stressed creating a large moment which could have buckled the truss diagonal?

We have one load case for the temporary construction where the supports were provided by the lifting trailers. After erected I can see the need to increase the jack forces due to the change in support conditions. When you look at those complicated joints I would not be shocked if they had some sort of bursting failure. It seems unlikely they could jack all the strands at the same time at some of those joints, and you could easily end up with a short term zone where you had a problem.

Looking thru the drawings I am confused when we needed to make our bridges into an art project? At one time the chicken merely needed to get to the other side of the road, but now it seems he needs to do so very elegantly. We could put in twice as many simple structures for what silly things like this cost.

I don't know why the won't release the actual video that shows the collapse. You can see a crew of 3 or 4 working on the canopy at the point where they tensioning equipment was located. When the bridge collapses, one of the workers it temporarily suspended in mid air. I wonder if he got hooked on the strand.

Quote (structuralengr89)

On Sheet 115/173..the first compression strut is labeled #11...It looks like there is no P.T. strand for that member...but there has been talk of that member Tensioned in the field???

The photos Ingenuity posted clearly show a tendon protruding from member #11. I'm assuming that was a change in the design sometime after that document in the pdf was created on 9/30/15.

Marco Rubio tweeted this last night: “The cables that suspend the #Miami bridge had loosened & the engineering firm ordered that they be tightened. They were being tightened when it collapsed today.” So it seems like it was either over-tensioned and the concrete crushed or a cable snapped and the jolt lead to a failure. The eccentric loading theory from tensioning sequence seems plausible too, particularly since based on the details in that pdf document, they'd have to go to the underside of the bridge to get to the other tendon(s) in the diagonal.

It's very obvious in hindsight they should have closed traffic for that since it's a heavily loaded compression diagonal, but they were touting their construction methods as causing minimal traffic impact, so I'm sure there was pressure to avoid any closures. This will probably be added to the list of failures covered in ethics courses or least to safety protocol for post-tensioning.

This just confirms something that Professor Henry Petroski (the author of 'To Engineer is Human: The Role of Failure in Successful Design' as well as other books on the subject) said once (I'm paraphrasing), "That eventually, particularly when designing bridges, all new types will be successful until the first failure is encountered, at which point we will learn what the minimum acceptable design criteria was, the last one built that DIDN'T fail."

His basic premise was that bridge design in particular was susceptible to a unique phenomenon and that is that since the vast majority of them are financed using public funds, that the pressure to continue to optimize the design is enormous, trimming away at safety factors and the costs of materials used, in order to win the bidding process, and that this will continue until we learn what constituted going just a bit too far. At that point, the consensus will be that the last successful project will be the standard by which all future ones will be be modeled. That is until a radical new design is introduced and the process will start over. In this case, it might not be the design of the bridge itself which was the victim of this phenomenon, but rather the method and processes used to construct it.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

John, while that's quite an insightful statement by Professor Petroski, I can't imagine that this particular showcase bridge in any way optimized public funds. Indeed, while skimming through the proposal the thing that jumped off the pages at me (as it was on, seemingly, every page) was how "environmentally friendly" the end product would be. When that issue is that much of a priority on a project designed to do one very difficult thing: defy gravity, well, in my opinion we've lost our way.

I believe what was optimized was political pressure; specifically, the pressure not to close down a road. It brings to mind a topic that was discussed on the structural forum, preventing a high-incidence / low-consequence failure (in this case preventing traffic impediments) leading to a low-incidence / high-consequence failure. And here we are.

Thanks to EngineerEIT for posting the proposal documents, which shed a lot of light on what was done. Not the construction drawings, but enough.

The way I read it, this was a concrete truss bridge with stays added for vibration control, or to look like a cable stayed bridge.

The proposal committed to an independent peer review. Wonder if that was done.

Concrete truss/frame, bad idea. Unbonded PT, bad idea. Very little bonded reinforcement, bad idea.

Wow, stays primarily for cosmetic effects? I did see that part of the proposal was to be able to have seasonal lighting projected onto them but it never occurred to me that that might be their primary purpose.

minerlax4"The photos Ingenuity posted clearly show a tendon protruding from member #11. I'm assuming that was a change in the design sometime after that document in the pdf was created on 9/30/15."


Are you sure about that? Could it be the tendon in the photo is actually Member #10...a tension member? Member #11 is under the greatest compressive force already...why would you add additional compressive force...if we assume this is a purely axial compressive force in a truss?...The only reason I would want to further compress this member, is if I thought the member had bending forces in it..and I needed additional moment capacity of member.

And the drawing shows that they were to apply the jacking force from the bottom of the truss...but the photo shows the stressing ram from the top...so the design is different than shown on the proposal (but that is not surprising).

I just read this article, which cites previous safety violations against Figg and MCM.

Link

structuralengr89, I think it was likely #11. On page 115 of that pdf, it shows the tendon arrangement for the panel point on the opposite end. If you mirror that, the #11 tendon, assuming #11 has tendons, would protrude in the spot that matches the tendon in the photos with the jack attached. It also appears to be pointing toward the end bent in the photos. I agree it makes no sense to be tensioning #11 when it's already under compression. Maybe you're right and it is #10, though. That would make the failure more difficult to explain, but would make a lot more engineering sense.

Quote (minerlax4)

Marco Rubio tweeted this last night: “The cables that suspend the #Miami bridge had loosened & the engineering firm ordered that they be tightened. They were being tightened when it collapsed today.” So it seems like it was either over-tensioned and the concrete crushed or a cable snapped and the jolt lead to a failure

I read somewhere where someone heard a 'snapping or pinging' sound a few hours before the collapse.

Does not bode well if the engineering firm ordered that the strand be tightened. Also, did anyone question why the strands loosened?

Dik

hokie66, FDOT is already pointing fingers about the third party review in this article. https://www.local10.com/news/florida/miami-dade/fi...

Quote (Archie264)

...while that's quite an insightful statement by Professor Petroski, I can't imagine that this particular showcase bridge in any way optimized public funds.

Please read the last sentence in my post, where I acknowledged that possibility.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

I suppose there is no stopping them, but it would be nice if politicians stopped putting out misinformation.

A rubbish article by an academic at Texas Tech. The guy is also on TV news, espousing his opinions. How do universities get sucked into hiring people like this?

He says the bridge should have had a centre support. Duh! That would have been a different bridge.

https://olivermcgee.org/fiu-bridge-collapse-why-mu...

Were those red shoring towers removed when collapse occured?

Here's a voice mail that the bridge engineer left notifying a stakeholder about significant cracks on the bridge.

Thought there wasn't a safety issue, although repairs were going to be needed.


https://www.google.com/amp/amp.miamiherald.com/new...

"We shape our buildings, thereafter they shape us." -WSC

tomfh,

Yes, it was spanning full length.

Quote (Ron)

Any possibility they tensioned this member out of sequence? Tensioning of this member should probably have waited for the cable stays..

Makes sense. Maybe they messed up the sequence, or maybe the sequence had errors meaning the strut had to carry the bridge and the stressing load - which was actually only intended to apply when the thing was hanging.

Quote (hokie66)

hokie66, FDOT is already pointing fingers about the third party review in this article

Pretty sad the way everyone bunkers down and tries to pass the buck rather than honestly appraising the situation. Reminds me of our professor's parting words: "When something goes wrong on one of your projects - and something will - the person who's fault it ends up being is the person who the blame ends up landing on."

No doubt the designer and builder will already have worked out why it's not their fault.

I hesitate to get involved but in some ways it is like the Shuttle O ring failure. The system didn't perform as expected. In the case of the O rings they leaked a bit on previous missions. In this case the tendons detensioned, the concrete cracked, THAT was the point at which everyone should have drawn a deep breath and start investigating. Trying to bodge it without understanding the root causes of the system's odd performance lead to this.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

I wonder if the last diagonal was more flexible / compressed more than they anticipated.

Compressing more may have caused the cables to loosen and tensile forces to happen where they weren't expected (the observed cracking).

The jacking the cables down more would have just added more load in that member, croaking the connection.

To the post-tensioners in the group- does that happen?

"We shape our buildings, thereafter they shape us." -WSC

Quote (Tomfh)

Tomfh (Structural)
17 Mar 18 01:14
Were those red shoring towers removed when collapse occured?

Yes, the portable red lifting rigs were only used to move the bridge segment from its (nearby) fabrication site to the installation point. They were NOT under the bridge when it collapsed - which had a full traffic flow: 3x lanes each way.

I have driven under that location several times while down in Miami for AWS and power plant work. Sobering to know I was at that exact traffic light - waiting to be crushed later.

Quote (greg)

In this case the tendons detensioned, the concrete cracked, THAT was the point at which everyone should have drawn a deep breath and start investigating.

Easy to say with benefit of hindsight, but concrete cracks, tendons detension, tendons snap, anchorages fail, and getting on with things is part of life.

No! When the bridge started making noises and cracks - that's when you stop the traffic, get people out from under the bridge and off of the bridge and try to get some sort of shoring under the bridge before it comes down. Even if you have to call the fire department! Most fire departments have some sort of training similar to Rescue Systems I which includes rapidly installing timber shoring.

It will be interesting to follow this investigation. One aspect of the project that I hope is scrutinized is the project delivery method of design/build.

Based upon what I have read, the engineering firm and the GC have worked together on several projects. Can this type of relationship lead to questionable decisions?

I am not making accusations, but I have been part of design build teams. The pressure exists.

This was just reported by CNN:

Engineer warned of cracking on bridge two days before collapse

https://www.cnn.com/2018/03/16/us/bridge-collapse-...

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

Quote (oldrunner)

No!

Yes, obviously in retrospect the traffic should have been stopped, the area cordoned off etc. But no-one knew that at the time. Presumably in their minds it was some ordinary cracking and misbehaving tendons, and they made that assessment, the same assessment all sort of engineers make everyday in response to issues - including engineers on this forum.

In this case it was the wrong call, no doubt.

That's assuming the cracking was in fact related. Do we know that yet?

Break down of video:

Frame Prior to Collapse:




Not sure if it is the video or if the first diagonal is buckled and maybe the section of top chord between the first diagonal and end of the bridge:




First diagonal appears buckled and the bottom chord has collapsed:




Following Frames:

EIT
www.HowToEngineer.com

Well actually done that on a few situations. One school building and a timber stairway at the same college. Evacuated building and blocked off the stairway. One building shifting and making noises. Told people to get out and the building collapsed two days later. One 200 stiffleg derrick - told superior that the derrick would collapse. Derrick collapsed and injured a guy. One building where the roof was failing - family out and building official red tagged the building. One unsupported crane boom boomed horizontal to hold a canvas with people under tables eating. Got the fire department to stop it. There's more but it's over a lot of years.

Sorry for the rant.

RFreund - Saw that yesterday evening as well but later when I looked at it, the video may have played differently. I'm wondering if the work on the canopy was to De-Tension Truss No. 11 on the north end. They may have been there for more than one purpose.

The original proposal had both sets of the Self-Propelled Modular Transporters spaced apart. Starting on the south end (FIU) the first rig was planned to be positioned directly under the south end of the span & the next rig 24 ft to the north. At the opposite end, north (Sweetwater)/Pylon, one rig was again to be positioned directly under the north end of the bridge with the final rig 27 ft to it's south. As it turned out this plan was too wide for the road way and they may have encountered a hiccup with uniformly supporting the bridge when driven over the center divider. It looks like they still had trouble driving over the center divider, even with the tandem Self-Propelled Modular Transporter arrangement. They stopped half way across and just sat their for a long time. Since the Self-Propelled Modular Transporter was no longer under the ends of the bridge, I wonder if they decided to PT the No. 11 Truss, with the intent to De-Stress the truss once the bridge was in place. It sure seems like it took a lot longer than 6 hours to position the bridge. It looked close to midday when they we finished. Link

If anybody still thinks this was really supposed to be a cable stayed bridge, you should review the drawings as to how the pipes were to be connected. They were to be bolted to the concrete with end plates. That is not how you connect cable stays.

Hokie66 -- exactly.

(Also, regarding that professor, absolute garbage writing. Unfortunately, it seems he's rather effective at self-promotion. Same way we get our politicians)

OhioMatt -- absolutely. Design-build projects inherently have more aligned interests, which can mean more trust and fewer checks. That's part of the efficiency, but can also be dangerous if people work outside their competency or back down too early in the face of peer pressure. It requires the individual engineers involved to be that much more careful to act ethically.

----
The name is a long story -- just call me Lo.

Quote (hokie66)

If anybody still thinks this was really supposed to be a cable stayed bridge, you should review the drawings as to how the pipes were to be connected. They were to be bolted to the concrete with end plates. That is not how you connect cable stays.

Quote (Lomarandil)

Hokie66 -- exactly.

And another "exactly" from me too.

16" CHS/tubes with bolt on connections - hardly your high-tech strand stay cables!

For those that wish to see the "PRELIMINARY 2015 - NOT FOR CONSTRUCTION" details, see the following:

"Easy to say with benefit of hindsight, but concrete cracks, tendons detension, tendons snap, anchorages fail, and getting on with things is part of life."

...said NASA management. Sorry, if you are designing stuff that risks people's lives then 'shit happens' is not an acceptable standard of care.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Quote (minerlax4)

On page 115 of that pdf, it shows the tendon arrangement for the panel point on the opposite end. If you mirror that, the #11 tendon, assuming #11 has tendons, would protrude in the spot that matches the tendon in the photos with the jack attached. It also appears to be pointing toward the end bent in the photos. I agree it makes no sense to be tensioning #11 when it's already under compression. Maybe you're right and it is #10, though. That would make the failure more difficult to explain, but would make a lot more engineering sense.

On page 115 of the submittal the PT bar schedule details ZERO PT for compression member #11, which I assume is for the final spanning condition:

Quote (epoxybot)

Since the Self-Propelled Modular Transporter was no longer under the ends of the bridge, I wonder if they decided to PT the No. 11 Truss, with the intent to De-Stress the truss once the bridge was in place.

I think this is most probable. During transportation/erection there was tension in member #11 so maybe they provided some temporary PT for this condition, and whilst this temporary PT was being de-stressed the collapse occurred. It is in keeping with the location of the stressing ram and de-stressing stool seen in the collapse photos.

May be a dumb question, but can someone explain why the top flange has any PT strands at all?

If the cable stays are just for show, when would the top flange ever be in tension to require PT?

So the tower and cables (and truss web pattern) is all simply decorative? That's pretty wild.

Quote (adamewood)

May be a dumb question, but can someone explain why the top flange has any PT strands at all?

For during lifting etc? as per diagram immediately above.

Also for bending of the top flange.

Quote (Greg)

Sorry, if you are designing stuff that risks people's lives then 'shit happens' is not an acceptable standard of care.

I'm not saying that. Just pointing out that cracks and tendon issues are pretty common in stressed decks and people don't generally hit the panic button unless it's out of the ordinary. Evidently this guy didn't think it was.

Quote (Tomfh)

For during lifting etc? as per diagram immediately above

Touche, sir.

By way of summary of the PT that made up this bridge:








The so-called media reports of "loose cables" having to be re-tensioned is highly suspect. All the photos that I have reviewed of the collapsed bridge show black-colored grouted end-caps (at the ends of the bottom and top flanges), and none that I have found to show any with 'stressing tails' exiting the ends, EXCEPT for the PT bar being de-stressed to diagonal member #11.

Ingenuity,
I don't interpret that as zero PT in #11, but rather as "ditto", so 280 kips.

I find all those numbers to be incongruous. Member #1 will have more force than Member #11 once the truss is erected, as it is not as steep.

adamewood,
Tension members are not the only axial elements which are prestressed. Think prestressed columns and piles.

Nice details ingenuity!

Quote (hokie66)

I don't interpret that as zero PT in #11, but rather as "ditto", so 280 kips.

It looks like zero to me. Same symbol as #1.

Quote (hokie66)

. Member #1 will have more force than Member #11 once the truss is erected, as it is not as steep.

Do you mean #2 instead of #1?

Quote (hokie66)

I don't interpret that as zero PT in #11, but rather as "ditto", so 280 kips.

If that was true then how much PT was to be applied to Member #1 - the first scheduled member and with a "-" for all PT data?

Quote (hokie66)

Member #1 will have more force than Member #11 once the truss is erected, as it is not as steep.

Member #1 is the end-vertical - I think you mean Member #2 - the opposite-end, first diagonal, that is flatter than #11. However, Member #2 [in the final 'faux' stay-cable configuration] is in direct line-of-action of the last stay cable [so may see some tension under some load cases ?] - whilst #11 is always in compression in the final configuration.

I did mean #2. Did you find reinforcement other than the PT tendons and bars?

Quote (hokie66)

Did you find reinforcement other than the PT tendons and bars?

No mild steel reinforcement drawings/details that I could find.

Over on The YouTube channel AvE, the host made an interesting point in his latest video (https://www.youtube.com/watch?v=KtiTm2dKLgU) - the tensioner was far out of the hole. His suggestion was that stored energy from tensioning popped it loose when the tensioning rod failed. If it was still pulling and the rod intact, it should be up against the concrete cap post-collapse.

This makes some sense as to the buckling of the concrete strut - the rods in the preliminary plans are off-center. If one fails and the other is still under tension there is a moment introduced into the member.

Maybe they were trying to pull hard enough to close a crack and popped one of the rods instead, leading to buckling of the member.

Per the original plan - there was no PT to be applied to #11. Not only is the tension "-", there is no member type (A, B, or C) specified to tell where the tensioners are to be placed in the member.

In this picture
it looks like a trench was cut on the anchor blister over member 10-11. This trench is not obvious in the post-collapse images. There is a similar trench into the anchor blister over member 2-3. Both align within the body of the most steeply sloped member. This is the same image as
the post epoxybot (Structural) 17 Mar 18 04:22 linked to; it's easier to zoom in on epoxybot's linked image.

Why the excellent video that was re-recorded on cell phone isn't available:

A county spokeswoman, Karla Damian, said the county's Traffic Management Center in Doral, which monitors more than 100 intersection cameras, found about the collapse through the media. Damian said it too late to preserve the incident before it was erased by new footage in the video system's rolling 30-minute recording loop.

"By the time that staff at the TMC attempted to access the video camera, it was already beyond the 30-minute threshold and our staff could not go back to the time of the collapse," Damian said in a statement Friday

http://www.miamiherald.com/news/local/community/mi...

This U Tube dashcam from the opposite direction shows it better, I think.

https://www.youtube.com/watch?v=Ucflj-MsJBI

similar driver footage. Hokie66 video above shows more detail

https://www.youtube.com/watch?v=BSq8295GFk4

Following from the youtuber theory that the rod snapping caused the collapse. If they were aiming to destress the rod in #2 and #11, why would snapping the #11 rod cause the member to fail? Was the temporary rod in #11 inadvertently holding #11 (and the node) together?

I am curious how this bridge was to handle unbalanced load effectively - i.e. all pedestrians looking over one side of the bridge? Seems it would not fare that well in torsion.

Quote:

I'm not saying that. Just pointing out that cracks and tendon issues are pretty common in stressed decks and people don't generally hit the panic button unless it's out of the ordinary. Evidently this guy didn't think it was.

"Didnt think" isnt an excuse when it comes to safety matters bc safety is process driven per engineering ethics. These circumstances should've been considered in a FMEA, circumstances outside the FMEA require the pause button being hit until the process docs are reviewed and updated. If this truly is a matter of someone "didnt think" then hopefully those folks end up in jail and out of the profession.

A possible scenario:
"Well Harry, how are we supposed to take the nut off that bar?"
"Nothing to it Joe. We use the hydraulic jack to stretch the bar a little and then spin the nut off by hand."

"Okay Joe, the nut should turn off now."
"Gosh Harry, I can't turn it. It seems to be stuck somehow."
"Okay. I'll stretch it a little more and see if it will turn."
"Oops. Too much."

Bill
--------------------
"Why not the best?"
Jimmy Carter

There should be two rods in member #11 if they duplicated what was used elsewhere. One rod, accessible from the top near the bottom of the beam and a second one accessible from the end that runs along the top. If the rods were both loaded and one became unloaded then the member is potentially destabilized. The state of member #2 is not considered. The rods in #11 should only have been needed for placing the bridge because the transporter lifted the bridge so that the end of the bridge was loaded opposite to its installed loading.

Thanks for the link. I had seen links to a dashcam, the one in the semi, but it was too low resolution, making the image a blur. In Russia there would might have been 40 videos. It's a shame the crane blocks the view of the anchor plate.

It is either reassuring (privacy) or disappointing (evidence) that the traffic-cam footage is automatically deleted. Good on the operators for having some presence of mind to record it before it was deleted. It had the advantage of not changing perspective, making frame-by-frame comparisons clearer.

I still don’t like the eccentricity that could be created in the web members by having multiple PT bars and no mild rebar. If a PT bar fails/is detensioned, there is a moment created on the diagonal by the remaining PT bar(s) and no mild bars to resist bending.

• I'm really disappointed with FDOT's Bridge Engineer allowing the release of the audio of the Engineer w/FIGG to the media to be broadcast. Maybe there are sunshine laws requiring the release of the info, but they could have fought that with their attorneys. They could have released a transcript of the voicemail. The audio does nothing to help in this situation, I feel for the Engineer and his family.
  
  
• Sure to be used in future Ethics courses...You have a client (FIU) who's claim to fame is accelerated bridge construction...which just means constructing a bridge to minimize downtime of traffic...And you have a problem with your construction...and really should shut traffic down to post-tension an extremely important member in you design...which brings me to the third point

• Lack of Redundancy...Most highway bridges in the US are pre-stressed beams or steel with multiple girders....beams are hit by trucks with cranes all the time...sometimes 1 or 2 of the girders are taken out..and the bridge still stands....this design had no redundancy at all...and I would be willing to bet that the design lead engineer and all the senior/junior engineers who did the calcs...were bitting their nails when you have a contractor up their stressing that joint...I know I would

The proposal is an incredible document. Very impressive.

I can't help but think that maybe there are things going on in a truss web/chord connection that maybe we don't quite fully understand. Did "graceful" top "robust" in this design?

I'm curious about the canopy. I wonder if the designer assumed the entire width was the top chord of the truss.

Just some ramblings:

This will be winding its way through the courts long after I’m gone. In addition to the fatalities, there are numerous other things. A lot of questions, so far unanswered. What really caused the failure; the video is poor and may not show the initial failure. Did the University really have an agreement with the FDOT regarding review? With construction going over a busy roadway, this seems reasonable. Was the FDOT aware of the construction; it appears to be widely publicised, and maybe a dozen others, but, not suitable for this forum.

If that voice phone message is correct, then, there may have been a precursor to the collapse that was not followed through. There is also a reference to a ‘pinging’ sound a couple of hours before the collapse. If acted on, several lives may have been saved.

I’m not a big fan of design-build because the owner is ‘short changed’ in the QA/QC of the project. Additional review may have been useful.

The cable stayed structure, from the renderings, appears to be an attractive structure. A lot of work went in to providing these asthetics. I don’t know why it was chosen. It is likely a very costly undertaking. I think steel would have been more economical and would offer greater ductility. The spans are not excessive. The depth to span appears to provide sufficient depth to use a 2-span structure, in concrete or steel, without cable stays. I’m not familiar with the quakes and/or hurricanes that Florida can expect. It may be that construction has to be more robust than what I’ve normally encountered. For hurricanes, I would consider that a massive, albeit ‘brittle’ structure, would have merit; I’m not so sure about a seismic environment.

I don’t know what the intended support at the ‘free’ ends of the cable stayed span. I couldn’t gather this from the drawings and it has a major impact on forces/stress throughout the structure. The sequence of stressing and relaxing the strand or bars has a major impact on the resulting stresses in the structure. As noted, some of the details don’t make sense for a cable stayed structure, in particular, how will it be ‘adjusted’ after erection.

[Added] I didn't know at the time of this post that the 'cable stays' were decoratively only.

I’ve taken a cursory look at the structural information. Some of it makes little sense. Not knowing the intent of the design, it may be that this could be rationalised. I just don’t know. As I noted earlier the compression force in the end diagonal may very well be a tension member in the final configuration. Just so many unknowns.

In these environs, a ‘dash’ in a table is regarded as ‘nothing’, not a ‘ditto’ or a “do”. Having no information in these spaces makes little sense, IMHO. It could be that the ‘dash’ means to repeat. One thing about-design build is that this type of notation may mean ‘whatever it should mean’ and can be easily interpreted. This reasoning applies to sections and details. ‘As Built’ drawings or record drawings may be a little different.

Will the city proceed with this work? It looks like the schedule is ‘shot’.

It irks me that an agency that employs video surveillance doesn't make use of screen recording software in the first instance and retrieval from the server in the second.

The bridge was built over Southwest 8th St. which is US Hwy 41. During the bridge move the site would have been crawling with FDOT engineers. The call from FIGG was simply good conduct to the Chief FDOT Engineer for the project.

Since the first line in the table of PT members has dashes, I would interpret the dashes to me NO PT.

According to the initial proposal, the bridge over Hwy 41 was essentially structurally complete when it was moved. The faux stay cables would add stiffness that would dampen harmonic ~play~ caused synchronous pedestrian foot traffic.

The project was a ridiculous amount of money for the FHWA to be spending on a pedestrian bridge. The entire idea of making this area into a public space was delusional. Who in their right mind wants to hang out next to 40 mph noisy traffic on a road that has little to none in the way of sound damping foliage? All the hard surfaces would have made it even worse from a noise perspective.

The bridge is, for all intents, on the back side of the university. There is nothing over in this area but parking garages. The city of Sweetwater is only 15000 residents. 74% foreign born, it is known as "Little Managua". When you do the projections, it really doesn't pencil out to spend this kind of money for a pedestrian bridge, so why?

Here is why: Link
A private development, that's why. This whole project stinks of politics, using Federal money to build a bridge for the benefit of private investors. The poor people of Sweetwater are going to see their city bulldozed in the next decade. The developer/s have already torn down the two story condos that were adjacent to the bridge, to build the tower. They call the tower "University Bridge" and it was originally suppose to be half condo half apartments. With the tax changes it is now to be all apartments.

They obviously would need a pedestrian bridge BUT if the university & Miami County wanted a signature bridge, then they should have been on the hook for at least HALF the cost, instead of the FHWA paying for 80% of the project.

What does 74% foreign born or "Little Managua" have to do with the pedestrian bridge?

Slickdeals... maybe the demographics didn't justify the bridge, and it was politically motivated.

Dik

I'm saying they are typically a less informed and vulnerable population that have founded a community that by City Data standards is a nice affordable low crime place to live. I also think that big developers and FIU with 54,000 students sees Sweetwater as ripe for the pickings.

I have not read all of the above comments. The attached video may have been mentioned before...one guy's explanation for the collapse.

https://www.youtube.com/watch?v=KtiTm2dKLgU

BA

I know we are all very curious to know what exactly happened, so here is my 2 cents.
I have very little experience with concrete, and none with pre/post tensioned concrete structures, but it appears to me from the dashcam video and the method of collapse that the first compression member's connection (where the alleged de-tensioning or tensioning operation was occuring) to the upper chord failed in shear.

If AVE's assertion is correct that the PT rod and ram being ejected some distance out of the conduit is evidence that the construction crew tensioned beyond the Fu of the rod, why would this cause a collapse of the entire structure? (being as it is a compression member and all)

What levels of redundancy are typical for a structure of this type?

Quote (structuraleng49)

I'm really disappointed with FDOT's Bridge Engineer

That's how it goes when things go badly wrong. You do someone else over so you don't get done over (that's an observation, not an endorsement). Evidently the strategy is working given the comments even in this forum here.

Quote (EPCISteel)

but it appears to me from the video and the method of collapse that the first compression member's connection (where the alleged de-tensioning or tensioning operation was occuring) to the upper chord failed in shear.

If AVE's assertion is correct that the PT rod and ram being ejected some distance out of the conduit is evidence that the construction crew tensioned beyond the Fu of the rod, why would this cause a collapse of the entire structure? (being as it is a compression member and all)

I had the same thought. Maybe the rods clamping stress was in fact keeping the member and joint confined, keeping the member and node intact.

Mr. "Va-Jay-O" in the youtube video linked by BAretired knows just enough to be dangerous, but he is seriously entertaining and has some nice tools at his disposal to boot.

Quote (3DDave)

There should be two rods in member #11 if they duplicated what was used elsewhere.

It appears to be a single rod hanging out the ends. That would make sense if it was just there to carry the weight of the overhangs during lifting.

Quote (epoxybot)

During the bridge move the site would have been crawling with FDOT engineers. The call from FIGG was simply good conduct to the Chief FDOT Engineer for the project.

I agree.

A meeting about the crack a few hours before collapse:

https://www.reuters.com/article/us-florida-bridge/...

Quote:

Engineers and state and university officials met hours before a new pedestrian bridge collapsed in southern Florida, killing six people, but concluded a crack in the structure was not a safety concern, Florida International University said on Saturday.

The meeting on Thursday involved FIGG, which is the private contractor for the overall bridge design, the school, Florida Department of Transportation officials and Munilla Construction Management (MCM), which installed the $14.2 million bridge.

EDIT: Tried embedding the video but I can't seem to get it. The link shows a blowup of the collapse.

<iframe width="560" height="315" src="https://www.youtube.com/embed/Ucflj-MsJBI?rel=0" frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>

"The cable stayed structure, from the renderings, appears to be an attractive structure. A lot of work went in to providing these asthetics. I don’t know why it was chosen. It is likely a very costly undertaking."

Here in the Dallas area, they have built some major bridges with big arches and whatnot. There's nothing below that would prevent them from putting piers down every 10' if they chose, and the older bridges were more along that line. I deduce that whether the big arches and cable supports serve a structural function or not, they were selected primarily for aesthetics, not because that was the ideal way to build something. I suspect the Florida bridge has similar concerns going on.

Tomfh (Structural)17 Mar 18 21:25

It appears to be a single rod hanging out the ends. That would make sense if it was just there to carry the weight of the overhangs during lifting.


Detail shows one pulled from the top and one pulled from the bottom....but it could have only been one in #11

Video from foreman with MOM stated that there were 2 tendons to detension (after bridge is set on abutment) and they can reopen traffic

I wonder if these 2 tendons were detensioned....traffic was reopened...cracking was observed so FIGG decided to retension the strands due to cracking...then when pulling the top one....it either snapped or the node busted

Quote (structuralengr89)

I wonder if these 2 tendons were detensioned....traffic was reopened...cracking was observed so FIGG decided to retension the strands due to cracking...then when pulling the top one....it either snapped or the node busted

That would make sense.

It would be interesting to see the photographs of the cracks that proceeded the collapse. I assume someone took photographs...

Dik

Quote (JStephen)

There's nothing below that would prevent them from putting piers down every 10'

For safety reasons you want supports to be well clear of the driving lanes. Clear span is good.

Dik

Quote (Dik)

It would be interesting to see the photographs of the cracks that proceeded the collapse. I assume someone took photographs...

Yep, there would be photos. I hope someone leaks that.

Quote (waross)

A possible scenario:
"Well Harry, how are we supposed to take the nut off that bar?"
"Nothing to it Joe. We use the hydraulic jack to stretch the bar a little and then spin the nut off by hand."

"Okay Joe, the nut should turn off now."
"Gosh Harry, I can't turn it. It seems to be stuck somehow."
"Okay. I'll stretch it a little more and see if it will turn."
"Oops. Too much."

That is a bit of a callous characterization, IMO. One member of the stressing crew died in the collapse, and two of his colleagues suffered injuries, meanwhile the engineer sat in his/her 'ivory tower'.

PT stressing operations are undertaken by experienced and specialized personnel, using calibrated and serviced equipment, and bridge projects even moreso (compared to building structures). VSL (the PT subcontractor on this project) have a worldwide history of post-tensioning that exceeds 50 years.

Don't run the field guys 'under the bus' - I know of several 'stressors' who have saved the engineer's a$$ a bunch of times.

I've read through all the posts and have a few thoughts.

First, these TIGER grants are always associated with some questionable projects. For example, back in 2009 the I-44 bridge over the Arkansas River in Tulsa, Oklahoma got a TIGER Grant that funded its replacement. The reason it got the grant? Because it was a MULTIMODAL bridge. It was a double decker bridge, the top deck for cars and trucks and the bottom deck for future light rail and also a pedestrian trail on the side. That the bridge needed to be replaced had nothing to do with getting the grant, it had the ability to support light rail in the future (that doesn't exist and will never happen by the way) and people could also walk across it. Therefore, it got the Federal Grant to be built. It was all politics.

This looks to be the same way. Link
Why in the world would you spend all this money ($11.4 million) for a bridge that would have tables and chairs with ceiling fans built into it? Elevators with glass walls. Solar technology, "eco-friendly" concrete, programmable LEDs. A large center pylon with fake pipe stays that don't serve any real purpose. This one span weighed 940 tons, how does it have a vibration problem? I don't buy that the pipe stays were for vibration, they were just there for looks. All of this just added to the cost. You could have built a much simpler steel truss (with redundant members) and still used ABC with SPMTs to set it in place and saved a lot of money.

Second, why in the world would you build a non-redundant concrete truss over 6 lanes of traffic? Too dangerous in my opinion.

From what I gather in all of these posts it looks like the contractor was doing something with the post tensioning on the diagonal member (#11) when the truss failed. I've never designed anything that was post tensioned, but my hunch is that they did something that caused that diagonal member to buckle. It's pretty clear from the video that P205 posted that the source of the failure came from that member.

Black humour... I thought it was funny and gave him a star for it... I can appreciate your sensitivity for the tragic collapse and loss of life, including the stressing crew member.

Dik

The video would suggest that we are focusing on the wrong member, 11. The collapse occurred at the top and bottom of member 10. And if others are correct, member 9 which intersected with 10 at the bottom, had no prestress. But wherever the failure actually occurred, the issue here is joints in a concrete truss/frame. In these forums, we often discuss problems to do with efficiency of concrete joints, and it is widely known that they can't be 100% efficient, but I don't know of any literature about design of concrete trusses per se.

The video that p205 posted was probably the best of the failure. Is there software that can show the individual frames and maybe enhance the image. The event happens so quickly I cannot determine if the failure started at the top of the diagonal member or at the base. The web member did not appear to buckle and it appeared that the panel point may have failed.

Dik

Here's a couple of frames from the vid. First frame is immediately before collapse. You can see #11 intersects with the node.

Following from that, something has happened to #11, and it no longer aligns with the node. Pity that crane is in the way.





Here's the relevant frames from the CCTV. Not a great video, however it appears something's happened to #11.

There are 6 canopy tendons altogether, but they appear to have two of them tensioned only for the movement portion and then relieve tension after placement. In pictures, 4 of the six are capped, leaving access to the remaining two. I believe those are what the guy in the video was talking about when he talked about removing tension.

The work on member #11 happens after cracking is reported.

I think the construction crew was working with faulty information and tried a fix for a condition they didn't understand.

Dash cam video relevant bit (I "dolly zoomed" so the bridge is same size in each frame):

I made a gif of 11 frames and cleaned it up a bit. Let's see if it loads. Well okay & the 10MB is in the attachments. I cleaned it up as best I can. After looking at the failure to the top left of #11 and to the bottom left of 10 & 9, I wonder if the position of the drains & the recess under the bridge for the drain pipe didn't create a point of weakness. Link

• http://files.engineering.com/getfile.aspx?folder=70ac024f-c25b-408d-9111-3e

Tomfh...great video w the dash cam...i see an explosion at the bottom of the vertical at the support...maybe the strand they were pulling at the top of #11 came out at the bearing node of the truss...the tension diagonal adjacent to #11 appears to punch thru at the bottom as well

Tomfh your cleaned up video is great! You should give copy to each party concerned. It looks like tensioning of the truss going down t the left pulled the top and the adjacent truss going down to the right and thus breaking the walkway then the top chord also broke.

Better yet secure copyright to the video and sell it to the parties. Delete your post here.

Thank you all for providing all thoughts and supporting documentation for this horrible accident. It is very obvious that de-stressing or stressing the tendon in #11 triggered the collapse. This diagonal should be completely de-stressed when the span was placed on the piers, but apparently it didn't happen. The cracking reported prior to the accident should result in emergency shoring of the bridge, as any crack in post tensioned structure should trigger such action.
The design of this pedestrian bridge was apparently driven by the appearance, not the structural efficiency and safety. The truss could serve without stays, but single diagonals and posts were not sufficient for the wind loads and required redundancy.
As usual, the devil is in the details - lack of mild reinforcement at the joints, where the stresses are almost unpredictable, post-tensioning of the compression member lead to this disaster.
Eugene Figg was a great engineer - but apparently the "successors" do not have the same capabilities. And even greatest legacy do not do the design - it's just the designer in charge of the project.
So, the recipe for for disaster is rather simple - great legacy company, inept design, and no response to apparent overstress signs.
Just bunch of idiots running the project.

Excellent and sobering discussion. It seems pretty likely that de-stressing member 11 caused the collapse. If the temporary PT was stressed to 200 kips, de-stressing it would have significantly increased the tension in the adjacent members. Listening to the voicemail was really chilling. I am certain that I have spoken almost the same words more than once in response to s**t happening during construction. Very often, the cracks we see really aren't safety issues. Engineers are human and can make mistakes. What kind of systematic rules could prevent this from happening again? Definitely no adjusting PT in non-redundant structures with traffic below. But in general how do we make sure the right decisions are made, especially during design/build projects?

Quote (graybeach)

What kind of systematic rules could prevent this from happening again? Definitely no adjusting PT in non-redundant structures with traffic below

How about not designing non-redundant structures...

We'll wait and see what the 'official' cause of the collapse was; hopefully it will be similar to this thread. It would be great to have copies of photographs taken of the 'crack', just to see how insignificant it was. The EoR was likely doing 'damage control' in the event others brought up the crack.

It will be interesting to see if the project goes ahead, and, if there is any 'political spin'. It would be nice to see if the funding is challenged.

I was really surprised that the collapse was 'instantaneous', without warning, other than maybe the earlier crack. The EoR will likely carry his statement about the insignificance of it to his grave...

Dik

This photo indicates that the deck slab which was acting as the tension chord of the truss, pulled away from the column and pier. The end diagonal, column and canopy slab still seem to form a triangle after collapse, although the connection between the end diagonal and column cannot be seen.

BA

• http://files.engineering.com/getfile.aspx?folder=014275c1-8eb3-439a-aff1-44

Reading through the above and (most of) the links, I can't see anything to say where the pre-failure cracks were found - apart from the "at the North end" in that voicemail message.

Have I missed something along the way, or do we not actually know that yet?

I wonder if the cracks might eventually turn out to have played no part in the ultimate failure. Too early to tell.

A.

BA, I had noticed the same thing... that the top chord, end vertical and first compression diagonal seem to still form a triangle after the collapse, but, as you point out, it's not clear what the condition or status of the joints is, or whether they are even still really connected or simply "leaning" on each other.

Also, in the screen shot that BA posted a couple of comments up thread, it looks like, in general, the PT anchor "blisters" on top of the canopy/top chord have mainly stayed in tact but fractured or sheared off the from the main canopy/top chord. That seems odd to me, unless they had been cast separately, had a higher f'c, had some mild steel reinforcing in them, or something along those lines. Anyone else think it is odd that the anchor "blisters" would stay in tact but separate from the top chord?

gte447f, true but there can be no doubt that the tension chord became disconnected at the end; and without the tension chord, the truss must fail.

BA

Wikipedia provides some details (with links to sources).

"An eyewitness reported that at the moment before collapse, a blue box fell loose from a crane hook, dropping onto the roof of the bridge very near where the roof and span then immediately broke apart."

BA, yes, definitely can't have a truss without a tension chord. Look at the zoomed in video posted by Tomfh up thread a little bit. It is very fuzzy, but in the second frame, it looks like you can make out something shooting up vertically from the top chord first panel point and in the same frame there appears to be an "explosion" of dust at opposite end of diagonal #11 at the bottom chord bearing end bearing. What we see in this frame could be the result of the PT rod rupturing under tension and erupting out both the top and bottom ends of diagonal #11. The object that appears to shoot up in the air from the top chord would be the stressing ram still attached to the live end of the PT rod. The dust cloud at the bottom of the diagonal would be the dead end anchor plate erupting out of the concrete. What do you think?

gte447f, sounds plausible. Perhaps the stressing ram is the mysterious "blue box" reported in the Wikipedia article.

BA

From this reddit post someone spotted two tendons in the PT Blister and if the mysterious "Blue Box" was the the control box in the same picture and it dropped, well..., that would have been very bad. Link

I'm skeptical of the "reports" of a blue box being dropped from the crane onto the canopy/top chord just prior to the collapse. Some of the same reports mention the blue box hanging from a blue cable. Who has ever seen a blue cable? Some of the same reports that mention the blue box falling from a blue cable quote congressmen as saying things like, and I am paraphrasing here, "they were conducting a stress test... they were impacting the bridge with a load to see what it could handle", implying they were intentionally dropping a load onto the deck to see if it could handle the impact. Of course that is absurd. Even Marco Rubio has made statements about the "cables that suspend the Miami bridge", when of course we know there are no suspension cables involved. So, there is a lot of uninformed information out there.

Now there was a crane there, and you do wonder why. What was its roll during the tensioning/detensioning operation that was going on? Could something have fallen from the crane? I suppose so, but why would something fall from the crane, and what would be large enough to contribute to the bridge collapse? What are the chances of 2 catastrophic failures (the crane and the bridge) happening at the same time, or I suppose one catastrophic failure resulting an an immediate second catastrophic failure?

Of the two blue things in the photos of the collapsed bridge, one (the cylinder) is the stressing ram/jack and is still attached to the PT rod which has clearly erupted from the diagonal about 5 or 6 feet, and the other (the box) I think is the hydraulic gage to monitor the jacking force. I think this gage would have been on the deck right next to the workers stressing the pt rod, not suspended from a crane.

Thinking out loud, all of the stressing equipment (hydraulic ram/gauge/pump) is heavy to lug around, and likely would have been hoisted up top with the crane instead of manhandling it up there. You can see maybe a lifting line and a red tether in the photo above that may still be attached to some of the equipment (I can't tell). Or, could it have been fall protection for the workers?

I apologize if my style is offensive to some.
I don't apologize for the basic content of my post.
I have seen too many versions of that play out over the years.
Sometimes with injuries, often with property damage and sometime with firings.
Who is responsible?
Sometimes the engineer in his ivory office.
Sometimes the engineer on the ground personally directing the work, possibly against the advice of experienced tradesmen.
Sometimes the crew is taking a shortcut.
Sometimes the crew does not follow directions.
Sometimes the crew does not understand the instructions.
Sometimes people get fired.
Sometimes the right people get fired but not always.
Sometimes the accident report is accurate and true.
Sometimes not.
And sometimes shtuff happens.
Why are cameras banned on many construction sites?
There is nothing like a photograph to screw up a perfectly good coverup.

Extraneous thought: After the Columbia bridge failure and now this one;
Maybe all cracks in important structure should be treated as serious until proven otherwise.
Inconvenient in many cases yes.
How much inconvenience is more important than a fatality?

Design/build?
Sometimes it works, but most of the old timers here have seen the results of profit driven decisions made on design build projects.
One that comes to mind is a pulp mill that was profitable for over twenty years.
Then came the design/build expansion and a newsprint mill was added.
There is nothing like improperly driven foundation pilings to mess up a newsprint drier.
In a few short years the company was out of business and the entire mill was dismantled and shipped to the third world.
The design/build company is still in business.

Bill
--------------------
"Why not the best?"
Jimmy Carter

waross... you sure you're not a structural guy?

Just curious...

Dik

gte447f - I have been skeptical of witness statements as well because they are often unreliable. There are actually 3 blue objects. The ram, the metering/pressure rig (control box) and the hydraulic fluid reservoir. With the arched canopy, the thought to sling the control box from the crane might have seemed like a good idea at the time. If indeed, this was the case, then the investigators probably already know that, when the control box fell and uncontrolled pressure surge to the jacking ram overloaded the tendon. Full Size Photo: Link

• http://files.engineering.com/getfile.aspx?folder=464bbac4-f6d7-4cd0-9612-dd

Original 'Powerteam' orange that was re-painted in VSL 'blue':



Full of oil, these pumps weight about 65 lb.

Stressing is always a two-person (minimum) operation: pump operator and jack 'stressor'. If you look at the videos that Tomfh and epoxybot post you can see a guy fall as his lanyard tie-off breaks, so likely that the pump and operator were in a 'dog box' suspended from the crane.

If the "pipe stays" were indeed almost cosmetic, then it bothers me a lot that all of the diagonals of the truss were aligned with the stays, instead of given simpler geometry.

Quote (epoxybot)

...metering/pressure rig (control box) and the hydraulic fluid reservoir.

I cannot see a 'third blue object'. And with a PE55 pump the 4-way control valve (for double-acting ram) and oil reservoir are all contained within the same box, as my photo above shows (the orange pump).

Good info Ingenuity.

Quote (AggieYank)

If the "pipe stays" were indeed almost cosmetic, then it bothers me a lot that all of the diagonals of the truss were aligned with the stays, instead of given simpler geometry.

Don't let some engineering get in the way of some bridge 'art'. It is in PINK for sarcasm!

Given the end result, where people lost their life, this is an inexcusable and unacceptable result.

Having said that, something to consider: If you have never worked for a contractor or on a project where you are responsible for project execution, then consider that something that appears black and white from the design office "ivory tower", is quite a bit grey-er in the field.

Or for political reference <G>

Dik

https://i.imgur.com/9o6zVT1.gif

Quote (It seems pretty likely that de-stressing member 11 caused the collapse. )

I agree, but the whole situation is still puzzling. Was the span designed as a truss or as an "I" section with web cutouts? If we assume it was designed as a determinant truss, Members 2 and 11 would have been in tension while the bridge was moved into place, based upon the position of the SPMT's in the photos, thereby requiring PT. Also,the deck or bottom chord of the first and last panels would have gone suddenly from compression to tension once the SPMT jacks were lowered. Perhaps in the final position, Member 11 had too much compression, which resulted in the "de-tensioning" operation.

IMHO, if it were designed as a truss it was the wrong approach. Forgive me if I'm rambling.

With regards to discussions about tensile failure of the PT bar, the stressing levels to PT bars were significant on some of the members, especially member #10.

The PT bar appears to be Williams Grade 150 ksi: Link.

For the 175' span truss, member #10 had its 4 PT bars stressed to 280 kips each. Min. UTS is 390 kips, so max was 72% of ultimate tensile. PTI (and manufacturer's) recommendations stipulate max of 80% of MUTS for test loads (usually for rock or soil anchor type projects where proof testing is undertaken), but there is also a requirement that the lock-off load should not exceed 70% of the specified minimum tensile strength. Threaded PT bar does not undergo significant seating losses like strand/wedge systems, so the stressing load is the lock-off load, for all intents-and-purposes.

The other PTI requirement is that the design load is not more than 60% of the specified minimum tensile strength of the prestressing steel.

The other PT bars to members of the 175' truss were stressed to 52% and 62% of MUTS.

I have not seen photos to show evidence of a tensile failure of the PT bar. The 10 foot 'projection' of the PT bar and jack from the canopy 'blister' could be explained by 'nodal' failure at the bottom of member #11 diagonal, where the bottom chord appears to have ripped away from the support 'node'.

What I see in the two photos are a Blue ram & control box with what looks like a blue plastic reservoir (with rounded corners) and then in the other picture a scratched blue metal ?reservoir?. Perhaps in the photo with the A & B designations the reason the third item cant be seen is because it is behind the fireman. IDK.

All the photos show me that the failure was a joint failure rather that a member failure. Both ends of Member 10 developed hinges at the same time. We have focused on PT in the members, but have no knowledge of bonded mild steel reinforcement in the structure, or of the joint reinforcement.

Just wondering...of all the engineers here, has anyone ever designed or had knowledge of a concrete truss?

Bridgebuster, you have it right, a truss made of concrete was the wrong approach.

NOTICE:

This thread is getting long. Discussion continued in new thread:

thread815-436699: Miami Pedestrian Bridge, Part II