Miami Pedestrian Bridge, Part XIII 81

[JAE]

A continuation of our discussion of this failure. Best to read the other threads first to avoid rehashing things already discussed.

Part I
thread815-436595

Part II
thread815-436699

Part III
thread815-436802

Part IV
thread815-436924

Part V
thread815-437029

Part VI
thread815-438451

Part VII
thread815-438966

Part VIII
thread815-440072

Part IX
thread815-451175

Part X
thread815-454618

Part XI
thread815-454998

Part XII
thread815-455746


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[IceNine]

The real tragic error in this case was the failure to realize the seriousness of the cracks and act decisively during construction.

Exactly. Looking at the photos from Eddy Leon the morning of the collapse, before the 9:00AM meeting, I don't understand how a PE could say there isn't a safety issue.

 

[Vance Wiley]

My take is you are right on point.
But consider if the whole thing were cast in place and falsework for both spans removed simultaneously, the thrusts from members 11 and 14 would have opposed each other and greatly reduced the net demand at node 11/12 (and then 13/14). Plus the large pylon section thru the truss would add capacity.
Basically the Stage 2 stand alone main span design ignored the fact that the back span would not be available to confine or assist node 11/12.
The difficulty emerged when the construction was staged to accommodate ABC techniques. I see the design intent as being to create one integral structure by common anchorage to the pylon and by placing continuity PT in the full length canopy. The continuity over the pylon also provided some sense of redundancy - I suppose that meant if one span failed the other could help it survive - or that both would then collapse.
I do not think the EOR intended that the remaining construction would close the cracking - that would have required pulling the two segments together over the pylon - something that I see being most difficult.
I think the EOR recognized at the March 15 meeting the only way to save this structure was to hurry up and complete the back span, perhaps add some tieback to "capture" node 11/12 until the backspan was complete, and then repair the cracking - probably with epoxy injection. I question the possibility of success with that. The cracking was extensive and member 11 was damaged.
Anything can be repaired, but with the damage evident on March 15 that would be a - " bridge too far "?.
And then the clock ticked midnight.

 

[Vance Wiley]

And wisely on the safe side. Too bad the designer of the bridge erred on the wrong side.
And if my friend the ME missed it, the toilet ran over. In the case of the FIU bridge, the consequences were horribly more severe.
Thanks - I had not considered the 50% factor.

 

[epoxybot]

IceNine (Structural) Maybe the CEI criteria needs to be updated to include certification of field staff for ATC-20/45 Training.

 

[Vance Wiley]

Thank you. The information may be in this link :

https://dms.ntsb.gov/pubdms/search/hitlist.cfm?doc...=[/URL]
The plans for construction are in documents #62 , #63 and #92
The calculations are in documents #94 and #95]

It took me a half hour to realize the numbers on the left side will jump you to the right place.
Once there I realize the model is broken into many many parts and locating a joint or member is difficult. Jump to section titled Main Span Erection Model on page 531 of the pdf file.
Now if there were just a map somewhere -
About 2 pages deeper there is an INPUT: Sections file with South end diagonal Area = 5.25 fr^2, a Pylon end Diagonal of 5.03 ft^2 (which equals a 21" X 34.5" dimension) and a yy shear area of 4.19 ft^2 and xx shear area same.
EDIT Note the term diagonal seems to apply to all web members.
That would seem to show the correct size was used in this analysis. Were these results ignored?
Can someone find the results of this analysis as they apply to node 11/12 and member 11?

 

[Vance Wiley]

As I recall, in the early 60s allowable stress in rebar was 20,000 psi, with (I think) 40 ksi yield.

From CRSI
1968
ASTM A305, ASTM A408, ASTM A431, and ASTM A432 withdrawn. ASTM A615 published (replaced ASTM A15, ASTM A408, ASTM A431, ASTM A432, and portions of ASTM A305) with grades 40, 60, and 75.

 

[Vance Wiley]

Fred - Thank you.
I can read my own approach to structures in your words. That is a comforting read.

 

[Hiway.1]

Regarding Figg's 2,000 or 6,000 or whatever pages of documents. That is a big part of the problem also. No EOR can provide adequate checking or oversight of all of that garbage documentation.

Same goes for the Figg's fragmented design team. You have a self identified stress analyst emphasizing that he is an analyst and not a designer. You have an engineer saying I only designed the substructure, you have an engineer saying I only designed the back span, you have an engineer saying I was the principal engineer for the superstructure but I just used the loads that so and so analyst gave me.

I think this hits the nail on the head for the first tragic mistake in this project. I don't think it got near the intentional design effort or professional oversight that it should have gotten. This type of design documentation would have never gotten past first base in a traditional FDOT design-bid-build process where district staff were doing Phase reviews on the project much less a Category 2 structure that has to undergo the supplemental review by Central Office in Tallahassee. Considering the uniqueness of this project featuring a Non-Redundant Reinforced Concrete Truss, how can so many professionals be asleep at the wheel?

Most of these thread's posts focus on the origin of the design error. I've designed enough of these types of projects to not be surprised by design errors; but rather expect errors to occur. What boggles my mind is the number of checks that the system has in place to ensure that these expected design errors don't make it into the wild and cause fatalities. The COMPLETE FAILURE OF THE ENTIRE SYSTEM is what shocks me here.

 

[IceNine]

IceNine (Structural) Maybe the CEI criteria needs to be updated to include certification of field staff for ATC-20/45 Training.

But Eddy Leon, PE is a Figg Bridge Engineer!

 

[samwise753]

There would be zero success with what the EOR thought would work. I have heard of remedial post-tensioning used to close cracks, but these are typically flexure cracks that have been arrested by flexure steel. They are just being closed for preservation against the environment. We're talking 40 times the acceptable allowable in this case; this was fractured concrete. Game over.

I would be skeptical of the front and back span approach regardless; I just wouldn't want to attempt to post-tension the two separate spans into continuity. The only way I'd even attempt that is if I moved both spans into place and kept them continuously supported until the continuity post tensioning was complete. The amount of force necessary to overcome the simply supported stresses would be staggering and likely overload the section in compression. Even then, I wouldn't want to do that; the structure is already non-redundant. If continuity post tensioning fails or loosens over time, the separate spans better be able to stand on their own.

 

[Vance Wiley]

I forgot to address this in my earlier reply. sorry.
I previously thought the EOR was referring to the entire backspan, but you may be correct. Realistically, the thru truss section of the pylon is the only thing that could be accomplished in the next few weeks.
Stage 4, Note 3 says "Cast intermediate section of the pylon". Note 4 addresses casting the deck.
I have assumed that section would be cast with the webs of the backspan and encompass member 13 and wrap member 12 but it is a specific step prior to casting the deck.
I would have posed that question in an RFI had the structure survived to that point.

 

[epoxybot]

IceNine (Structural) - Understood, but the first party to witness the cracking and who represents the FDOT on an LAP project is the CEI. When BPA's inspector saw the cracks on the day the bridge was set, BPA failed to use the depth of their Pre-qualification's staff to identify what they were seeing. They also failed to generate crack maps to communicate effectively with all parties. They failed to suss out from their PT inspector, on site at the time the cracking occurred, if the cracks happened before or after detensioning.

The worst cracking happened BEFORE detensioning. They were the only party who could have conveyed this information and they failed to even identify the actual event that mattered to all that followed. Everyone went into the meeting on the day of the collapse believing the detensioning had cause the extensive cracking. Figg would never have proposed retentioning if they had known when the cracking actually occurred. Maybe they even would have been more inclined to ask themselves, what they might have got wrong and closed the bridge.

 

[Vance Wiley]

Totally agree that pulling (or pushing) south to reset node 11/12 would not work. The cracking has created chunks in the cracks and it just wont go back. Leaving the slip in place would cause problems elsewhere. And a force horizontal at 32 degrees to support and raise a structure this heavy is not a good idea. So it would need to be shored and lifted.
Then they could chip out the damaged deck area (hoping the D1 PT did not explode its anchors) and remove the damaged member 11 and rebuild it in place. A lot of work on a new structure.
I wonder if someone would have stepped back and just said no?
Had someone said no, the design engineer could then have claimed he was being prevented from restoring it to perfect condition.
Before that became a viable possibility, the clock ticked midnight.

 

[IceNine]

Epoxybot,
I agree that BPA could have done a much better job, and that prior to arriving on site on the 15th Figg’s decision to retension may have seemed reasonable. But after viewing the cracks on the morning of the collapse, it should have been back to the drawing board and road closed.

 

[epoxybot]

IceNine (Structural)
Absolutely but the bridge should have REMAINED closed the day it was set. Particularly since it was ABC and had just been set. The cracks should have been identified as structural immediately and the FDOT notified before the roadway was reopened. The fact that there were existing cracks and subsequently additional cracking then propagated from the same zone was reason enough to keep the roadway closed and wait for the FDOT to make a determination.

 

[hokie66]

I think some of the cracking initiated even before they lifted it off the ground. Restraint shrinkage cracking. The diagonals tried to restrain the deck, and of course that was impossible without cracking.

 

[Kestrel42]

Vance, I’ve spent all day trying to find the dimensions of member 12 that FIGG used for their models, and I’m having no luck confirming that what MCM said in their analysis is correct. However, I think we are on the right track about the pylon/member 13 pour happening prior to the backspan. If you look at the construction plans for the pylon, it indicates a cold joint vertically between the pylon/13 and the the backspan elements of the deck, member 14 and the canopy. Not how I originally pictured the construction sequence but that seems to have been the plan.

The final slide from the EOR’s presentation on March 15th states as the final conclusion:

The spalled areas are minor and it is recommended that they be prepared using normal procedures and poured back along with the up coming “pylon diaphragm” pour (different from and prior to the back span on falsework pours).

Methinks the EOR was hoping the temporary shim and PT bar stressing in 11 would keep things in place until they could get this pylon diaphragm stage completed. Since the pylon would wrap around the side of 12 (and those 4 exposed pipes on the side of 12) it would have presumably have helped 12 resist the shear force from 11 pushing against it.

The real question I have now is which member 12 configuration(s) were used for the computer modeling of the main span. Did they use 12 alone, as at the time of the collapse? Or was it 12 as incorporated into the pylon diaphragm as MCM claims? Considering how many elements differ between the models and the as-constructed main span, it seems like FIGGs models were even more of a mess than we thought.

Pylon plans:

https://dms.ntsb.gov/public/62500-62999/62821/628545.pdf

FIGG March 15th presentation:

https://dms.ntsb.gov/public/62500-62999/62821/628510.pdf

 

[Vance Wiley]

Where are you finding the Pylon details? The Index says they are on drawings B-22 thru B-26, which I cannot locate.
The only indication I can find is Detail 1 on B-55 and Half Plan on B-57.
Found it - the Pylon is #92. Plz xcuse.
I am going to add comments as part of this post.
I do not see a joint in the added concrete at the sides of member 12. These wrap-arounds reach 2'-10&1/2" across the full side of 12.
Detail E-E is at the 8" pipe below the deck level and has a cold joint to the end of the deck, B-B appears to be below the deck with a joint to the end of the deck, and Detail D-D is above the deck and does not indicate a joint in the concrete as it wraps member 12 on the east and west face of 12. Perhaps the EOR was going to issue a revised detail.
A joint at the end of 12 would leave only that portion of the new concrete atop the deck, and would be difficult to mobilize that area of the deck because it is already cracked badly. More dowels into the top of the pylon could be used, but the force from node 11/12 is 4 feet above that surface.
I give little credit to the ability of the concrete pylon alone to help, let alone halt the failure. But that may have been the intent on March 15. We can add that to other bad decisions made that day.
Also casting the interior pylon to the top of the pylon below before the deck is placed blocks the reaction of member 14 and prevents its transfer of vertical and horizontal forces to the support at the south end of the north span. I think they are borrowing more trouble than they could have designed their way out of if they cast the interior pylon section before the north span deck.
One more add. Section A-A, Drawing B-49, shows the top of the mid support "Pylon" with the main span diaphragm 2, "pylon filler" (my word, 2 ft wide) and 2 foot wide diaphragm 3 of the north span. There appears to be a pour joint between the "Filler" (my word) and the diaphragms 2 and 3. No way that 2 foot filler would have resisted the slide from node 11/12.
Plus it must stop moving so the concrete can set.

 

[jrs_87]

FIGG Bridge Engineers statement regarding the NTSB board meeting of October 22, 2019

https://flbridgefacts.com/downloads...s_statement_regarding_ntsb_hearing_102219.pdf

At the NTSB meeting today, it was evident that the investigation into the FIU pedestrian bridge
construction accident presented challenges for the agency to accurately understand all of the technical
and factual components. The accident was the result of a complex series of events and failings by parties
at multiple stages of the project.
An analysis conducted by Wiss, Janney, Elstner Associates (WJE), the country’s preeminent forensic
structural engineering firm, proved that if the construction joint at member 11 had been built as required
by Florida Department of Transportation (FDOT) Standard Construction Specifications, the construction
accident would not have occurred.
WJE has worked with the government on numerous forensic investigations in its 60-year history such as
the 2007 collapse of the I-35W Mississippi River Bridge in Minneapolis. In its review of the FIU pedestrian
bridge accident, WJE conducted detailed research, in-depth analysis, and physical testing. When the
Federal Highway Administration (FHWA) Turner-Fairbank Highway Research Center issued its October
19, 2018 analysis concluding that the concrete cold joints were not intentionally roughened, WJE went a
step beyond analysis and performed testing of full-scale replicas of the critical connection with both
roughened and un-roughened surfaces. WJE's tests revealed, contrary to the findings of the NTSB, that
the failure to roughen the concrete beneath bridge member 11 was the fundamental cause of the
collapse.
FIGG Bridge Engineers, Inc. looks forward to an opportunity to discuss WJE's findings with FHWA.

 

[Vance Wiley]

It is interesting that WJE did not test a joint that was EDIT "roughened in a manner" END EDIT to clean and remove laitance but without "intentionally roughened to 1/4" amplitude".
The FIGG statement is supporting their directive to the project to prepare the construction joints to FDOT requirements and claiming the WJE testing confirms the capacities of "FDOT joints".
As was pointed out by that is a "bait and switch" call - the FDOT reference does not require a " 1/4" amplitude".
WJE tested "as placed" and "intentionally roughened to 1/4" amplitude" joints. We still have no idea how a joint prepared to FDOT requirements would have performed. And they used different sources for cements and certainly different aggregates in the testing, thereby requiring a "correction factor" in their results to predict the performance of the aggregates used in the actual construction.
That seems to leave questions on the table.
But I can agree with this part - -
EDIT ADD FDOT says "FOOT Standard Specifications 400-9 States: 400-9.3 Preparation of Surfaces: Before depositing new concrete on or against concrete which has hardened, re-tighten the forms. Roughen the surface of the hardened concrete in a manner that will not leave loosened particles, aggregate, or damaged concrete at the surface. Thoroughly clean the surface of foreign matter and /aitance, and saturate it with water." (Copied from FIGG paper presented to NTSB).
From prior comments on this forum I get the idea that the application of this FDOT requirement varies from do nothing to do a bit more. That is borne out by responses in interviews by NTSB.
Because FIGG directed the jobsite to use FDOT specifications, they have defined the standard for this work.
The important point for FIGG is will this preparation according to FDOT perform as well as a "1/4" amplitude" preparation and therefore merit a coeff of friction of 1.0 instead of 0.6? Proper testing could have addressed this point directly.
Because WJE found the maximum resistance was at a slip of 0.02 to 0.025 inches, a joint properly meeting the FDOT spec might have performed as intended, warranting a coefficient of friction of 1.0 . That could have been valuable to FIGG. The conspiracy part of me has to wonder if that was tested but unfavorable results were not reported.
Does anyone know of an instance where this FDOT requirement was tested in concrete of this strength level?
Is there a "standard understanding of intent" regarding this requirement?