Teach me please

[Paul_D_North_Jr]

Not on-line, but this looks like what you want: (in FORTRAN, no less!) 

https://trid.trb.org/view/38423 

See also this: http://armytransportation.tpub.com/TI-850-02/Figure-7-19-Cooper-Load-Configuration-For-Bridges-112.htm 

I have to confess, the "floor" part of this puzzles me.  The structural trilogy I remember is "shear - thrust - moment".  I can see having the moment and shear values - they can be relatively easily calculated from just the span length and loading.  But for the floor beam reactions you'd need to know their spacings - at the panel points for a truss, but they could be anyplace for a girder span. 

- PDN. 

 

[Euclid]

VOLKER LANDWEHR

Here is an interesting summary of the fact known about the bridge failure of 03/30/2018: http://www.miamiherald.com/news/local/community/miami-dade/west-miami-dade/article207358659.html

They state that external structural engineers have offered explanations in engineering forums but acknowledge that the lack of knowledge final structural drawings and analysis still makes them speculations.
Regards, Volker

Edit: While browsing through the enginering forum linked in the above article I found the following super slow motion version of a dash-cam video: https://www.youtube.com/watch?v=D6pGzgm4zZ4&feature=youtu.be

For me it looks like the support diagonal failed at the joint with the top chord leaving all load carrying to the bottom chord which failed in bending fracture. Still only a description and no reasoning.

 

The article points out the principle of structural redundancy that is often incorporated into bridge design.  The FIU bridge lacked redundancy due to its reliance on a single truss.  Trusses inherently lack redundancy, but if the bridge just had two trusses, which is a common design, it would have had the redundancy of the second truss.  The FIU bridge would have had plenty of redundancy if it were a cable stayed suspension bridge.  But ironically, while it was built to look like a cable stayed suspension bridge, that appearance was just for show; just to give it class and make it look majestic. It was not a structural feature.   

[ChuckCobleigh]

Paul_D_North_Jr

Not on-line, but this looks like what you want: (in FORTRAN, no less!) 

Yikes!  I think it's been nearly 30 years since my last FORTRAN decks were written; it would take a serious sweeping of cobwebs to go through FORTRAN code now, I am guessing.  (BTW, at least at that point the "decks" were actually computer files, not real decks of Hollerith cards. For some reason, though, we still called them decks.)  Did have about 13-14 years of FORTRAN, though, before moving on to deal with Pascal, C/C++ and (wince) Ada.  Don't miss any of it now, not the least little bit.

[blue streak 1]

Do we assume the failure occurred at the original location of the center pylon ?  Remove that pylon and all structure engineering needs to be redone .  lenghting any planed  span with out complete re engineering is a receipt for disaster  -------  

[tree68]

Paul_D_North_Jr

(in FORTRAN, no less!) 

Took FORTRAN when I was going for my associates in business - ended up writting a program that required almost an entire box of cards (which I had punched).

Went back later for data processing - learned Cobol (two versions - one for PCs), RPG, and Pascal and a few other applications.  I've learned Basic on my own, and still do a little work with it.

The dangers of being a "command line commando."  I can do some HTML, too, although some of the newer stuff escapes me.

We've come a long way from the IBM 370 I did FORTRAN on to the mighty machines we carry in our pockets now that probably have more processing power and memory than the mainframes of that era.

[Euclid]

blue streak 1

Do we assume the failure occurred at the original location of the center pylon ?  Remove that pylon and all structure engineering needs to be redone .  lenghting any planed  span with out complete re engineering is a receipt for disaster  -------  

 

The center pylon planned location was moved northward and that is where it was built.  Then the southern bridge span was placed with its northern end on the center pylon.  The failure apparently occurred at the connection of the northernmost diagonal web member with the span canopy/top structure.  At that connection, the steel tension members were accessible from the top of the canopy, and adjustment of the tension was underway when the collapse occurred.

You are right that the original plan had to be revised before the center pylon was built.  This required major redesign including the following:

1. Move the center pylon 11 ft. north; redesign the pylon and its foundation for the change in ground elevation and closer proximity to the water canal.

2. Lengthen the main truss section by 11 ft.

3. Shorten the second truss section by 11 ft.

4. Change the angle, length, and location of all 20 diagonal web members.

5. Change the angle, length, and location of all 10 cosmetic members representing “cable stays.”

I assume that the design work needed to successfully make the above revisions was done, but it added more time required to complete the bridge. Experts have pointed out the inherent risk from making assumptions about how much re-checking needs to be done when redoing the design and engineering work.

I think the most plausible theory about the cause of the collapse is that the first truss section was damaged during the installation process using the large, mobile jacking tables. 

Like the bridge design itself, this installation procedure needs to be carefully engineered and designed.  As I understand it, this installation procedure was changed very recently, perhaps only days before the installation.  The original plan called for supporting the truss in four places, and that was revised to supporting in two places.  In so doing, the support at the north end was placed many feet south of the north end of the truss, whereas previously, there was a support point right at the north end. 

The theory:  The revised lift point, inset from north end, was located where the truss was less able to carry the loading of the lift.  And the lifting at that location was accomplished with multiple hydraulic rams that were operated independently without automatic equalization.  These factors damaged the truss structure by overstressing it.  The cracks that were observed and reported after the truss had been set in position may have resulted from this overstressing.  However, we have not been told were these cracks were located and what they looked like. 

If this theory is true, it points to a failure to perform the proper engineering and design for the truss installation. 

[Anonymous]

Euclid

Experts have pointed out the inherent risk from making assumptions about how much re-checking needs to be done when redoing the design and engineering work.

With today's structural analysis software I see this possible danger much smaller than in the past.

Much larger is the danger some intermediate states of construction might have been missed. I read somewhere that jurisdiction ruled that the consulting engineer only owes the analysis for the final state not the intermediates. But I might have misunderstood.

In Germany the Fee Structure for Architects and Engineers defines the scope of engineering. Intermediate states have to be contracted additionally.

If something went wrong with the analysis of intermediate states of construction, and I say if, that might be a design fault but not necessarily by the primary designer (Figg).

Euclid

And the lifting at that location was accomplished with multiple hydraulic rams that were operated independently without automatic equalization.

Usually multiple hydraulic rams are equalizised. I'd wonder if someone had taken that unnecessary risk.

Your theory contains too many ifs and uncertainties to discuss it in detail. So only two questions why did the bridge stand 3 days in its final position before failure if overstressed and why did the bridge survive post-tensioning work at the south end and not the north end?

What wonders me is the timing of the final post-tensioning work. For whatever reason this was needed I would have done it while the bridge rested on the transport supports and part of the load transferred to the final supports, longer traffic disturbance or not.

But that is hindsight. I believe the building crew felt safe with what and when they did it.

As the linked article summed up there is much information needed for a sustainable theory. Without this information there are dozens speculative theories possible.
Regards, Volker

[Electroliner 1935]

tree68

Paul_D_North_Jr (in FORTRAN, no less!)  Took FORTRAN when I was going for my associates in business - ended up writting a program that required almost an entire box of cards (which I had punched).

Brings back memories of 1963, Took a course on Fortran, one on Microwave Systems, (our company was installing a system), and taught a course titled "Basic electronics for Generating Plants" for the employees at a generating station. One night while doing this, I went to bed and I couldn't get to sleep. I was programing the microwave and diagnosing the plant electronics. The mind would not shut down. Would look at the clock and it would be five minutes since I had last looked at it. One of the worst nights I have ever experienced. 

And as you said, the computing power in our pockets still amazes me. In the 80's the company was leasing 56 kBs data circuits, Now my homes' internet speed is 75 MBs. My cellphone shows 37 MBs on my router's wifi and 45 MBs on its 4g connection. My grandmother lived before men flew. Wonder what my kids will live to see. 

[rdamon]

At a recent tradeshow our company showed 400GB Ethernet interfaces.

Still to small for some search engine and social media companies. :) 

[Euclid]

VOLKER LANDWEHR

 

 

 

Euclid

And the lifting at that location was accomplished with multiple hydraulic rams that were operated independently without automatic equalization.

 

 

Usually multiple hydraulic rams are equalizised. I'd wonder if someone had taken that unnecessary risk.

Regards, Volker

 

Somebody mentioned that the high speed video of the installation showed the individual hydraulic cylinders extending and retracting out of unison; seemingly at random; apparently to jockey the truss into position; and apparently subjecting the truss to momentary localized load spikes as individual cylinders assumed more than their equal share of the loading. 

I recall seeing such a video, but can’t find it now.  In any case, I am curious about equalizing the cylinders.  It seems like it would require each cylinder to be fed by its own individual injector pump with the multiple pumps perfectly synchronized.   

Then, not only would the cylinder lift be equalized, but the cylinders would have their extension synchronized.  With these attributes, the level of the bridge deck during the lift and transport would depend solely on the road surface supporting the tires on the lifting platforms.  So that road surface would have to be level below the bridge if the bridge was expected to be set in a level position.  All that the hydraulic lift could do would be to raise or lower the truss span, but not change its level attitude established by the level roadway. 

[Anonymous]

Euclid

With these attributes, the level of the bridge deck during the lift and transport would depend solely on the road surface supporting the tires on the lifting platforms. So that road surface would have to be level below the bridge if the bridge was expected to be set in a level position.

Most modern self-propelled modular transporters are height and load compensated, the ones used in Miami sure are: https://youtu.be/R_1nogbOLQ0?t=56

Or here crossing the median: https://youtu.be/i2GAndUZcCY?t=16

Look at the wheels and you see that they are at different heights. The SPMT compensates the road roughness.

The movement I realized was moving the SPMTs back and forth to get it in the correct position.

Here is another video. It shows the lowering to the supports starting at 0:45. Otherwise I've seen no movement of the hoist cylinders.

https://www.msn.com/en-us/video/fun/timelapse-shows-construction-of-fiu-bridge-that-later-collapsed/vp-BBKgQkW
Regards, Volker

[Euclid]

VOLKER LANDWEHR

Most modern self-propelled modular transporters are height and load compensated, the ones used in Miami sure are: https://youtu.be/R_1nogbOLQ0?t=56

Or here crossing the median: https://youtu.be/i2GAndUZcCY?t=16

Look at the wheels and you see that they are at different heights. The SPMT compensates the road roughness.

The movement I realized was moving the SPMTs back and forth to get it in the correct position.

Here is another video. It shows the lowering to the supports starting at 0:45. Otherwise I've seen no movement of the hoist cylinders.

https://www.msn.com/en-us/video/fun/timelapse-shows-construction-of-fiu-bridge-that-later-collapsed/vp-BBKgQkW
Regards, Volker

 

I would think it would be ideal to carry equal loading on all of the truss span support points during the entire move.  Also, it would seem essential throughout the lift, move, and set-down, to maintain the exact bridge deck plane that will exist once it is in place.  So the support cylinders would adjust elevation as needed to conform to uneven elevation in the road surface. 

The video you posted does appear to show all of that, including compensating for irregular ground as the wheels move across the median.  I don’t see the cylinders acting independently to adjust the bridge level as it was being prepared to be set down on the piers.   

I seem to recall another video that showed the cylinders individually adjusting in an up close view from under the bridge, looking up at the cylinders from maybe 30 feet away.  But I can’t find that video.  I did not pay much attention to it at the time. In some of this recent reference material, there was someone in a video who also referred to that independent movement of the cylinders as being shown in a video.  It would be nice to see that installation video again. 

From the recent article that you posted:

“The engineer, W. Denney Pate, said the cracking did not pose a safety hazard but would have to be repaired. Cracking is common in concrete structures and is often superficial, but can sometimes be a sign of deeper trouble.”

How was it determined that the cracking did not pose a safety hazard?  How was it determined that the cracking was not “a sign of deeper trouble” as the article mentions as being a possibility with observed cracking?  It seems to me that since this cracking was formally acknowledged as a possible sign of structural danger, there would have to be a clear explanation of how this was found not to be the case. 

[Anonymous]

Euclid

I would think it would be ideal to carry equal loading on all of the truss span support points during the entire move.  Also, it would seem essential throughout the lift, move, and set-down, to maintain the exact bridge deck plane that will exist once it is in place.  So the support cylinders would adjust elevation as needed to conform to uneven elevation in the road surface. 

This is taken care for by the SPMTs. They can be programmed that the bridge's plane doesn't change even when crossing a median.

Euclid

How was it determined that the cracking did not pose a safety hazard?  How was it determined that the cracking was not “a sign of deeper trouble” as the article mentions as being a possibility with observed cracking?  It seems to me that since this cracking was formally acknowledged as a possible sign of structural danger, there would have to be a clear explanation of how this was found not to be the case. 

First, by definition, concrete is a material that can crack under load because of its low tensile strength. Therefore it gets reinforced with steel. The calculation is done with a tensile strength =0. So cracks are daily business for concrete designers. 

To determine how problematic a crack is you need to know the structural analysis and have a lot of experience.

I don't see an easy way to explain the reasoning behind cracks to laymen without going deep into concrete design. If I would read such an explanation, why this particular crack is not dangerous, I would have to believe it. Without knowing the dimensioning and not having seen the crack and its location personally I see no chance to judge the conclusions.
Regards, Volker

Edit: Sentence in Italics added to clarify the reference

[Paul_D_North_Jr]

"+1" to Volker's post immediately above, especially the "a lot of experience" part.

- PDN. 

[Euclid]

Just to be clear, my question is not meant to say that I require absolute proof that the cracking was not a symptom of a structural failure that posed a safety problem.  I understand that this sort of proof can be very difficult to demonstrate to a layman.  All I am asking is whether such proof was established.

The news article says this:

“The engineer, W. Denney Pate, said the cracking did not pose a safety hazard but would have to be repaired. Cracking is common in concrete structures and is often superficial, but can sometimes be a sign of deeper trouble.”

“Does not pose a safety hazard” is a decisive statement that means that absolute proof was established.  The statement does not include a qualifier that it was given as an opinion.  Yet we are told that very intensive investigation will be conducted to find the cause of the collapse, and that is far from complete, despite the fact that the above statement has been made.

Therefore, assuming that the article has quoted my Pate correctly, are we to understand that it is an absolute certainty that the cracking was not a symptom of a structural failure that would pose a safety hazard?

Or is it possible that Mr. Pate was only offering his professional opinion without absolute proof? 

[Anonymous]

When I was called to judge cracks in concrete I have seen my answer if the cracks were harmfull or not as my  professional opinion backed by my knowledge and experience . I was able to explain why I thought this way but there is mostly no proof, just knowledge and experience.

I never thought of the engineer's statement as a proven fact. For me it is his professional opinion backed by his experience and his knowledge of the structure.
Regards, Volker

 

 

[Euclid]

VOLKER LANDWEHR

I never thought of the engineer's statement as a proven fact. For me it is his professional opinion backed by his experience and his knowledge of the structure.

Well what about the ultimate structural performance and reliability of the bridge?  With all of the engineering and testing that goes into the bridge design, the final assurance that it is structurally sufficient is surely not just a professional opinion, is it?

 

[Anonymous]

A structure is analyzed compliant with respective building standards. These standards were developed from research and development and contain the knowledge at the time of its implementation.

Standards are not carved in stone. They develope over the years as the knowledge about in this case concrete gets better. During my career I learned to work with 4 generations of concrete standards. All steps allowed a better utilization of concrete. Sometimes innovative developments like high performance concrete get standardized.

And if you find weak points, mostly durabilty, you have to correct the standard. The influence of road de-icing salt on concrete in parking garages was underestimited till the end of the 1980s.

But I can't remember a change necessary because of a reduced safety.

And then there is the human factor.

So yes, if done correctly a structure is safe and that isn't an opinion.
Regards, Volker

[Euclid]

If bridge integrity is to be a matter of proven fact rather than just an opinion;

And if all they had to assure the safety of the FIU bridge until the cracking could be analyzed was an opinion;

Then I must conclude that keeping the road open below the FIU bridge during the analysis of the cracking was a terrible act of negligence.  

[Anonymous]

If it was a good idea to leave the road open while post-tensioning single truss members is questionable.

From my point of view there was no reason to close the road on account of a crack that was judged as not dangerous by an experienced engineer.

These decisions are not made thoughtlessly. You would check the analysis, look into reinforcement and post-tensioning drawings, try to find the reason for the crack and find out the crack's influence on the load carrying ability.

I don't know of any non-destructive testing to proof harmlessness of cracks. But I'm retired for 10 years now. You can find rebars, crack planes, but how and if the cracked structure carries the loads still remains the judgement of an engineer. For the proof you expect you have to cut out the cracked area and test it in a laboratory.

That is why I said, for judging cracks you need knowledge and a lot of experience.
Regards, Volker

 

 

[BaltACD]

Yes but - And their off!

[Euclid]

Well, we don’t know whether the dismissal of the cracks being a safety issue was based on a professional opinion –or- on a proven analysis.   So I will give them the benefit of the doubt, and assume that it was based on a proven analysis.  I make this assumption based on the fact that they did not close the road. 

 

This would seem to be quite a feat of investigation and analysis because not closing the road would mean that the proven analysis had to have been completed in just an instant of time in order to not leave the road open during the resolution of the question.

 

But from Mr. Pate’s statement, I conclude that when the full investigation is finished many months from now, we will be told what caused the collapse, and we will learn that the observed and reported cracks just prior to the collapse had nothing to do with the cause of the collapse, as either a symptom of a structural problem or any other relationship with the cause of the collapse.    

[Euclid]

Euclid

If bridge integrity is to be a matter of proven fact rather than just an opinion;

And if all they had to assure the safety of the FIU bridge until the cracking could be analyzed was an opinion;

Then I must conclude that keeping the road open below the FIU bridge during the analysis of the cracking was a terrible act of negligence.  

 

Cracks where FIU bridge buckled may have signaled 'imminent failure'

http://www.miamiherald.com/news/local/community/miami-dade/article210449384.html

 

"The cracks are telling them that the connection is failing, but they’re not seeing it," Howell said. "When it’s your design, you rationalize your way around it."

 

Finally we are seeing close up photos of the cracks that were discovered and dismissed as not being a problem.  It was said that cracks appearing in concrete are common and normal.  When I look at that crack in the first photo, I would have a hard time believing that it is normal.

Apparently the bridge integrity, after it was questioned due to the discovery of cracks, was just an opinion; and not subjected to a conclusive engineering analysis; as the bridge sped on to complete its mission of validating Accelerated Bridge Construction (ABC). 

[Anonymous]

I would just advice to wait for the NTSB report. Cracks are very difficult to judge without knowing the structural analysis in detail.

Yes the crack looks quite and it needs attention for corrosion prevention. This photo alone doesn't tell anything. I couldn't tell where on the bridge the crack was and if it is from the crashed bridge.

The FIU bridge looked like a truss bridge. While on steel truss bridges the truss nodes are calculated as freely rotatable, they are rigid on the concrete FIU bridge.

If you don't include this rigidity into your structural analysis or the reinforcement design you get cracks and a freely rotatable node. But the structural integrity is still guaranteed.

I would expect that the cracks were examined. If the following judgement was correct or not the NTSB report will hopefully show

Euclid

Apparently the bridge integrity, after it was questioned due to the discovery of cracks, was just an opinion; and not subjected to a conclusive engineering analysis;

We don't know what was done to verify if the cracks were harmless or not. In many cases you don't need any anylysis. Detailed knowledge of structural analysis and reinforcement drawings is often enough to determine the cause of a crack and its severity.

The Accelerated Bridge Construction has nothing to do with speed of construction. I find the term misleading. The bridge is built in normal speed beside the final site to avoid interrupting the traffic over the whole construction period.

ABC limits the traffic interruption to the time needed to move the bridge from the building site to the final site. If there had been problems with the crack found before moving the bridge it could have been fixed without endangering ABC.
Regards, Volker