Genoa Bridge Collapse: The Road to Tragedy | Hacker News Reader

Genoa Bridge Collapse: The Road to Tragedy

nytimes.com

238 pointsconroy125 comments

Bridges have lifespans and need constant maintenance.

One of my first tasks as an intern at the Army Corp Of Engineers was to create 6 inspection packets (copies of every part) of the cape cod canal bridge, which was being inspected.

I know the tappan-zee bridge was replaced because the old one was just too old (designed to last 50 years, it was replaced at over 60 years).

https://en.wikipedia.org/wiki/Tappan_Zee_Bridge_(1955–2017)#...

One of the reason I left the engineering profession is no one wants to pay for it so it gets "deferred". The US's bridges get a c+ according to the American Society Of Civil Engineers.

https://www.infrastructurereportcard.org/cat-item/bridges/

I'm also a little curious why the cables were wrapped in concrete. Steel cables are common on bridges. stability side to side? Concrete is not great in tension (steel is very good in tension). As the article indicates, it was potentially hiding some of the cable corrosion, making the situation worse.

>> I'm also a little curious why the cables were wrapped in concrete. Exposed steel cables are common on bridges. stability side to side? Concrete is not great in tension (steel is very good in tension). As the article indicates, it was potentially hiding some of the cable corrosion, making the situation worse.

the concrete was meant to be a protective sheath to stop cable corrosion. poor materials engineering at the time sounds like the issue

>> He believed using the system would reduce the sway of the bridge. Structural engineers seemed to agree.

But Mr. Morandi also believed that the concrete coating would protect the steel cables inside from the wear and tear of the elements.

“Concrete structures seemed to be eternal,” Mr. Majowiecki said. “This was the mentality.”

In that hope, he added, Mr. Morandi was greatly mistaken.

The concrete of the day turned out to be highly vulnerable to degradation, worsened perhaps by salty air from the Mediterranean Sea and the harsh fumes from nearby factories.

Cracks in the concrete shell let water in, and the steel began corroding almost as soon as the bridge was opened for traffic in 1967. But unlike bare cables, any corrosion was hidden deep inside, making it hard to detect."

Thanks,

Cracks in the concrete usually mean bad concrete. In the sites I worked on we didn't use much concrete, but the few times we did, we had to do a onsite slump test and send cylinders off to the lab to testing after curing. If the mix is wrong it effects the strength significantly.

One of the things about Civil Engineering as a field is the huge responsibility in designing things that when they work nobody notices.

Same thing with 90% of software engineering. Your work is notices only when it fails or sucks.

Concrete has become ubiquitous, cheap, and increasingly poured by people (including professionals) who don't appreciate, let alone understand, the chemistry and art. Walking around San Francisco and other cities I see brand new concrete (sidewalks, buildings) with cracks adjacent to 30-year-old flawless concrete.

Someone attempted to parge the concrete foundation of my house (before I bought it), and it simply popped off. Reason? Everybody has switched to impermeable cement (e.g. Portland, Quickcrete, etc) these days, but my foundation (like most from that era) used permeable cement. So moisture and salts travel through the foundation, reaches the parge coat, and the pressure pops it off. I didn't understand any of this (including what parge meant) but it only took a few hours of bored Googling one night, when I was contemplating redoing the parge coat myself, to begin to wrap my head again things.

My neighbor had his entire foundation replaced. They lifted the house and I guess they did a good job. (I wouldn't really know!) I was chatting with the supervisor and decided to ask him if it was worth properly parging my foundation, and what to look for to know whether I was getting the right kind of materials that will work with my foundation. He had NO CLUE what I was talking about. Maybe I didn't understand the lingo and had some facts wrong, but even if so I'm pretty sure someone knowledgable would have realized what I was trying to ask. But, nope. The guy is a professional who specializes in concrete--maybe not an engineer but someone laying foundations which need to be structurally safe--and just... clueless. All he knows is a particular process, not the whys, not the alternatives. And much like with consumables, the culture is to simply replace, not fix. It's cheaper to replace than fix because that's what the industry is optimized for.

That's where we're at, today. Expertise is gone and expectations have suffered.

Umm. Why would you want to parge coat that foundation? Did you try to do that on the outside or the inside?

Do you have concrete slab against ground? Cellar? Or is there unused airspace between the floor and the ground?

My 1926 house, like most in SF, has an unreinforced strip foundation. (Concrete in my case, which supplanted masonry no later than 1900-1910, AFAIU.)

The foundation is structurally fine (home inspection passed, earthquake retrofitters didn't seem bothered by it even though in at least one spot a installing foundation bolt blew through the concrete), but it's noticeably disintegrating. I presume the previous owner attempted a parge coat on the internal walls to hide this. My motives are both esthetic and practical. Practical because, theoretically, a proper parge coat can act as a sacrificial layer, slowing the disintegration. Whether a parge coat would actually extend the useful life of the foundation as practical matter is something only someone with real in-depth experience and knowledge of SF construction would be able to tell me. I just don't know how to find such a person.

If you tour SF homes[1] you'll notice that most have remodeled the garage to include one or more bedrooms. The strip foundations are trapezoidal--slightly wider than the sill plate at the top (which is about a 1 or 2 feet above ground), expanding to much larger as it goes down 3 or 4 feet under the ground. Older remodels often parge and paint the foundation or box it behind wood molding; newer ones might actually hide it behind the refinished walls. But because of the awkward dimensions, whatever they do you can easily spot it if you pay attention.

What I find crazy about this is that they're enclosing foundations which are (or will be) nearing their useful life. (SF homes are technically detached, but usually only have a few inches of gap at the most so the side walls are inaccessible.) Moreover, the refinishing is probably wrong. Either it'll pop off, or if they used a chemical binder they're (theoretically) accelerating the disintegration (the moisture and salts will create pressure behind the binder, pulling on the face of the older concrete). Likewise for impermeable[2] paint. Yet you won't even notice any of this because it'll be completely hidden.

I'm happy that the home we found was (all things considered) both relatively well-maintained and unmolested. The garage was never refinished except for a parking slab poured circa 1970s. I'd like to keep it this way.

[1] Which typically sit over a garage with a parking slab poured decades (or even a century) after the foundation. The garages were serendipitous. Most of the homes were built before cars were so common; what are now invariably garages were unfinished first floors intended to be used and finished (if at all) however the owner wanted--storage, workshop, in-law, extra living space, etc. When cars exploded most were turned into garages. Now in-laws and extra bedrooms are becoming much more common, either replacing the garage or reducing it to a size that would barely fit a lawn mower. Occasionally you'll find the odd house where there's still just sand at the first floor.

[2] Permeability of paint is a whole 'nother can of worms. Like the foundation, the typical stucco finishes on homes on the west side of SF were designed to be semi-permeable. The external walls need to breath. But idiot painters (including whoever the previous homeowner hired) commonly use a low-permeability latex paint. (Latex comes in all kinds.) Moisture accumulates behind the stucco and the paint quickly bubbles and peels. In some cases (like ours) it can be bad enough to cause moisture damage inside the house because vapor barriers were either non-existent or primitive (our house was wrapped in tar paper), and in any event both the inner and outer walls were designed to be semi-permeable as compared to modern construction.

Of course, ask most painters about these permeability issues and many or most wouldn't really understand what you were talking about, even in SF. Contractors who do a good job often do so because they've inherited a workable process (proper types of paint, etc), not because they understand why. But often times you still need that deeper knowledge to be able to debug or solve certain issues. Good luck finding somebody capable of doing that. I think that, like with modern network security, the saving grace is that most people end up dodging bullets by replacing whole systems before the consequences are realized.

Is the concrete seemingly leaking some white powder? If it is, then you have something to gain from external non permeable layer. Something like this?

https://www.rakentaja.fi/keskustelukanava/img/20157/295425.j...

I live in Finland. Here if the walls of a room are not completely above ground, you risk molds and therefore it's never advisable to have living quarters in such rooms.

If you do have living quarters in such rooms, the proper way to do it is to install non-permeable membrane between the soil and the concrete. The renovation looks like this: https://www.vastavalo.net/albums/userpics/13316/normal__MG_1...

If it's just a garage or storage space with no white leaks, I would just wait it to crumble to a degree that justifies changing the whole foundation. I don't see how internal coating could help anything. And external coating helps only very little if it's above ground.

I'd guess SF is so arid that people don't really notice that stuff. It's different in a country like this that has swamps and rain all over.

There's a little efflorescence, but nothing out of the ordinary for the area; it's more uniform and nowhere near as severe as your picture. I suspect installing a barrier wouldn't make any sense. Because of how houses (and foundations) abut, largely limiting access to within the property's footprint, I imagine it would be easier to simply pour a new foundation. New foundations are increasingly common, but I'd rather not drop $30,000+ before I need to.

SF is coastal and foggy. Most of the western half of SF is built on sand dunes. Drainage is excellent but the ground is nonetheless perpetually wet. One of the things I've always noticed about typical first floor remodels is that they always feel very damp. They may be fancy and expensive but they always feel like jail cells to me. We recently toured a house around the corner from us with a 1990s- or 2000s-era first floor remodel and, indeed, you could both smell and see mold. Unless you pour a new, modern foundation (or somehow manage to install an external barrier) I don't think it can be avoided. It's crazy to me that people pay premiums for such remodeled houses. Like I said, I'm glad the house we found was relatively unmolested.

You're probably right that the reasonable thing to do is leave it alone and replace it when it gets too far along. But then I'd have no excuse to learn an old-school masonry skill. :)

Sounds like the water comes just from capillary action. That is very good situation compared to actual ground water pressure pushing moisture through the walls.

You should be good if you keep there good ventilation and avoid putting any wood dust or wood structures directly in contact with the concrete. Wood and concrete combined are breeding ground for dry rot and that fungal nightmare can tolerate more arid environments than any other.

One trick you could try is to put some asphalt outside right next to the wall. It won't solve any underlying issues, but it can help a little bit.

I think the crumbling comes from something else than moisture in your case. Probably oxidation of cement accompanied with temperature variations. You can treat that in some cases by injecting cement into the cracks in the wall. But that is very expensive.

  I think the crumbling comes from something else than moisture in your case. Probably oxidation of cement accompanied with temperature variations.

Interesting. Thanks!

Not at all to try giving any lesson on the matter, but using membrane is a modern (and cheap) way to solve an issue that the Romans some 2000 years ago had solved in a much "cleaner" way.

In Italian it is called "scannafosso", this is the "modern" (and cheap and "wrong") way:

https://www.coffeenews.it/wp-content/uploads/2010/02/scannaf...

and this is how it was done traditionally (as said since Romans times) and "right":

https://www.coffeenews.it/wp-content/uploads/2010/02/scannaf...

The idea is that the walls must be permeable and areated, the membrane keeps the outside rain/water/humidity outside BUT it also keeps the internal humidity/vapour inside which is a good start for moulds and similar, and in any case it is definitely not healthy.

BTW the effect is not entirely dissimilar to what happens (hopely happened as nowadays it is more rare that new houses are built without appropriate ventilation) in modern houses built along "energy efficient" rules, any house in an energy class better than D (C, B, A and A+) without a mechanical air ventilation system is simply and plainly unhealthy for its occupants.

>> Not at all to try giving any lesson on the matter, but using membrane is a modern (and cheap) way to solve an issue that the Romans some 2000 years ago had solved in a much "cleaner" way.

...Proceeds to give lesson :P

If I'm understanding it correctly its basically a space dug down and left empty to keep the walls ~60cm away from the foundation? In most places, but especially SF, there isnt space for an extra meter on each side of the houses

>If I'm understanding it correctly its basically a space dug down and left empty to keep the walls ~60cm away from the foundation?

Exactly.

There is no actual need of 60 cm, which is only a "sane" width to allow inspection, the external wall and "air chamber" can be much smaller, see as an example:

http://docplayer.it/docs-images/41/9277133/images/page_8.jpg

and if there is not enough space for the above (25-30 cm) a wall of porous/hollow bricks (thickness 10-15 cm) will still do better than waterproof membrane alone.

>In most places, but especially SF, there isnt space for an extra meter on each side of the houses

I know, and I am not at all saying that it is a universal solution, and - besides - it costs a lot more than plainly applying to the house foundation and earthed walls this or that kind of non permeable membrane (there are bitumen based, PVC based, synthetical/chemical liquid compounds, bentonite, etc.).

Still if you are building a house and have the needed space, it is money well spent, I have seen tens or hundreds of recently built houses using membranes notwithstanding the availability of space (because it is easier and cheaper) with humidity/mould problems that cannot simply be resolved without mechanical ventilation (besides when needed repairing the waterproof layee(s)).

That was interesting, thanks.

But you might be pleased that the most modern system at least in Finland actually has little of the same features than the romans had. If you look at the same membrane I posted earlier, it has little knobs that hold it little bit away from the wall:

https://monosilla.files.wordpress.com/2014/11/screenshot-201...

I don't know the English word for it. It's not perfect, but lot better than couple decades ago. They used to paint the concrete with tar. This "patolevy" is cheaper, easier to install and allows little bit of ventilation.

Yes, I know those, though - technically - they are not a waterproofing membrane, they are intended (at least here) to be only a mechanical protection and a separation layer, and the height of the knobs is anyway too small to have an actual ventialtion.

Besides - usually the "top" of the layer is not properly sealed/protected so - over time - the space (at the bottom)is filled with fine sand/dust brought by wind/rain, of course preventing any ventilation where it is more needed.

"Bigger brothers" do exist, like:

https://www.youtube.com/watch?v=dnxj9mTN9NE

still they are usually placed over the waterproofing layer, this should guarantee:

1) that the actual waterproofing is not punched by a stone or whatever when filling the excavation

2) that the waterproof membrane is not "immersed" in humidity

What happened here seems to bear some similarity with Evans Hall (home to Math, Stats, & Econ, among other units) at UC Berkeley: https://en.wikipedia.org/wiki/Evans_Hall_(UC_Berkeley)#cite_...

Beside the (wrong) idea of protecting the cables from corrosion, the concrete, as you said, was there to improve stability.

It is pre-stressed concrete, it is working in compression against the cable tension. The idea is that as the load increases, instead of pulling directly on the cables, it just decompresses the concrete.

You can think of the cable as a spring, with the bridge suspended it. As you walk on the bridge, it bounces. To prevent the bouncing, you can put the spring under high tension into a rigid pipe, with the ends of the spring attached to the ends of the pipe. This way, as you walk on the bridge, the spring tension will stay constant, only the compression on the pipe will vary.

It is pre-stressed concrete, it is working in compression against the cable tension.

This seems strange to me.

My understanding is that pre-stressed concrete is concrete with reinforcements inside that apply a compressive force on the concrete. As mentioned by GP, this is to mitigate the effect of tensile forces within the concrete since concrete performs poorly in tension.

Are you suggesting that the cables within the concrete are not just stays, but also the component of the pre-stressed concrete applying a compressive force? Or is the reinforcement some other component?

I would assume that if the cables themselves were applying compressive force to the concrete, then their use as stays releases some of this force since the stays are in tension. Since concrete itself performs poorly in tension, it's going to provide a negligible counter to reinforce the cables even if it is still compressed, since the compression forces are provided by the cables. It seems to me that putting pre-stressed concrete into tension via its reinforcements is defeating the purpose of having pre-stressed concrete in the first place.

Or is concrete in some compression able to respond better in tension by weight than say, a thicker cable at the same weight as the given cable and concrete combined?

>It seems to me that putting pre-stressed concrete into tension via its reinforcements is defeating the purpose of having pre-stressed concrete in the first place.

I think you are correct there. But it looks like the stays were in separate tubes with little allowance around them to separate them from the pre-stressed concrete.

That airspace around the stays then would prevent the concrete from providing any chemical protection against corrosion. And concrete as physical protection against rainwater is often only as good as the paint in the surface of the concrete. It's often way too porous even without cracks.

> I would assume that if the cables themselves were applying compressive force to the concrete, then their use as stays releases some of this force since the stays are in tension. Since concrete itself performs poorly in tension, it's going to provide a negligible counter to reinforce the cables even if it is still compressed, since the compression forces are provided by the cables. It seems to me that putting pre-stressed concrete into tension via its reinforcements is defeating the purpose of having pre-stressed concrete in the first place.

Not quite. High resistance steel is used in prestress applications, which provides a considerable headroom to determine what cables are used (number of strands per cable group and number of cable groups per cable) and how to design a structure (target prestress values, dynamic/vibration properties, load variation/fatigue design, etc). Therefore, just because a designer selects a specific prestress cable that doesn't mean the cable will be designed to be stressed closed o yield or even fatigue limits. The designer can pretty much design a structure that uses prestress cables loaded at a fraction of their design limits but subjected to higher loads at construction to be able to unload the initial prestress load as compression. This strategy is often used in bridges, as is designing a bridge to e intentionally heavier so that variable loads are a fraction of the dead load to increase fatigue life.

Well yeah. But typically that is done by cross bracing. Lighter and more effective against the leverage of the whole bridge.

Concrete has low Young's modulus, so you need very wide concrete casing to get the same benefits in stability you would get from steel truss of similar dimensions.

It's incredible how well documented and rich of information is the article. I'm Italian and I could not find in any Italian newspapers all the information that I needed! It makes me aware of how poor is the information in my country!

JFYI (Italian) among many others, these are the more technical ones:

https://www.ingenio-web.it/20966-il-crollo-del-ponte-morandi...

https://www.ingenio-web.it/20925-il-viadotto-sul-polcevera-e...

https://www.ingenio-web.it/20939-ing-camomilla-come-e-perche...

True, but those are technical ones. I couldn't find any article explaining what a "strallo" is, in fact "cos'è uno strallo" (what is a straw) seems to be a common search based on Google's autocomplete.

>True, but those are technical ones. I couldn't find any article explaining what a "strallo" is, in fact "cos'è uno strallo" (what is a straw) seems to be a common search based on Google's autocomplete.

It's not a "straw", it is a "stay" in English.

Strallo is a term mutuated from boats, a sail boat has a mast (albero) that is kept vertical and strong by one or more sets of "stays" (stralli):

https://en.wikipedia.org/wiki/Stays_(nautical)

https://it.wikipedia.org/wiki/Strallo_(vela)

On the sail boat the function of the stay is "opposite" to that of a bridge stay, it is used to keep the mast pressed to the base and more rigid, on a bridge they are simply "suspension supports" allowing to carry the weight of the viaduct deck and transferring the vertical load to the tower or pylon (torre o antenna).

https://en.wikipedia.org/wiki/Cable-stayed_bridge

https://it.wikipedia.org/wiki/Strallo_(ingegneria)

I share the sentiment

The best piece I've read about the disaster

> a structural engineering professor at the Politecnico di Milano, Carmelo Gentile, found troubling signals of corrosion or other possible damage...he has performed his tests on some 300 bridges around the world

It would be very interesting to know what he had to say about the other 299 bridges. After all, you can find someone who will tell you you have a structural problem with the safest bridge. It's hard to tell if they are a crank, someone with a vested interest, someone seeking publicity for their new diagnostic method, or genuinely an expert who is giving you good advice. But if Professor Gentile mostly says the bridges are fine, but singled out this one as a danger - then it's really worth listening to his other recommendations.

Since he was contracted to do the inspection, I presume they found his diagnostic method credible.

Slightly off topic, but they way they changed the info graphic as you scrolled was _really_ cool! This is one of the rare cases where extra JavaScript makes sense - it really improves the experience.

Thanks, now I have to click the link ;-)

EDIT: the effect is really quite cool and actually useful unlike other animations!

NYT is famous for their beautiful yet informative scrolling infographics:

https://www.nytimes.com/interactive/2018/01/27/technology/so...

I found the opposite, that article is really irritating to read, I gave up a few pages down.

By "cool" do you mean it helped you read the article? I'm curious because I had the exact opposite response, I actually quit reading the article at that point. I think it might work better on a small screen, but I was reading it on a full size monitor with the window maximized. It actually took more time to scroll to the next sentence than it took to read the sentence. I found it too annoying to concentrate on the content.

How does this entire article get written without a single mention that much of the road infrastructure here, bridge included, had been privatized and was being managed by the Atlantia corporation. The article actually seems to put the reader on the mistaken impression that Autostrade (former name of Atlantia) is some sort of media management company.

"Autostrade, which handles media queries on behalf of the subcontractor, Spea Engineering, declined to comment."

https://www.cbsnews.com/news/italy-bridge-collapse-genoa-atl...

https://en.wikipedia.org/wiki/Atlantia_(company)

The technical specifics of the bridge collapse are interesting, but the privatization of multiple road systems should have in theory let the company be cognizant of the systemic issues of the design practices of various eras, but how was this missed? The company was managing the bridge since 1999, almost twenty years now. This is as much or more of an organizational question as well as a private vs public ownership question as it is a technical one.

You read a different article than me:

> He warned the company that manages the bridge, Autostrade per l’Italia, or Highways for Italy, but he said that it never followed up on his recommendation to perform a fuller computer study and to outfit the bridge with permanent sensors.

> For reasons it has not fully explained, Autostrade, which took over management of the bridge in 1999, did not carry out the same operation on the supports of the other two towers — including the tower that collapsed.

> Autostrade won the concession to run nearly half of Italy’s highways from a cash-strapped Italian government, starting in 1999. After that, there were no major renovations of the Morandi bridge.

... etc ...

Why are they using the wrong name of the company? I think I missed it because I was looking for the Atlantia name. Looks like I did miss quite a few references that way!

Looks like Atlantia is a conglomerate that has a large controlling interest in Autostrade per l'Italia and similar businesses. Autostrade per l'Italia is still a business entity, so it seems appropriate to me for the article to refer to it.

http://www.atlantia.it/en/operations/italian-motorway-operat...

edit softened snarky language.

But why then treat Autostrade like a pr management entity?

"Autostrade, which handles media queries on behalf of the subcontractor, Spea Engineering, declined to comment."

I guess that's what bothers me a bit - is that there is a somewhat in depth investigation regarding the technical explanation, but a big shrug at hitting the wall on explanation of management failures, which IMHO should go all the way up thru the ownership conglomerate.

I don't read that sentence as saying that Autostrade are solely a PR company (and certainly the rest of the article makes clear that this is not the case), just that they are managing talking to the media for that subcontractor.

I feel like you formed a conclusion based on your incorrect knowledge of the name of the company. Now that's been corrected, and your error highlighted, you're doubling down rather than admitting you made a mistake.

That sentence stuck out when I read it, too, but the article makes it clear that Autostrade was not just a PR co.

That's not strictly the wrong name. Autostrade per l'Italia SpA manages the highways and is owned by Atlantia, which was formerly named Autostrade SpA.

I'm honestly and sincerely wondering if you read the same article.

"Autostrade won the concession to run nearly half of Italy’s highways from a cash-strapped Italian government, starting in 1999. After that, there were no major renovations of the Morandi bridge."

> of the road infrastructure here, bridge included, had been privatized

That's not correct. They're not privately owned: they're a sort of lease (I can't quite translate "concessione") because the government is still technically the one who onws it.

And for those other HN readers, there's also a public-owned company that handles roads, Anas, which is not doing well with regards to maintenance (this just to put into perspective that's not just a public vs private thing).

> (I can't quite translate "concessione")

I think in English it's "concession" :-) "The right to use land or other property for a specified purpose, granted by a government, company, or other controlling body"

> by using investigators’ descriptions of a central piece of evidence — video footage captured by a security camera.

Whenever I read about these incidents, with only one camera angle, I almost want to buy a camera and point it outwards from my apartment just in case I catch something like this happening in my city.

An artist in New York took a picture out his window almost every day for 11 years with a large-format film camera. He indeed caught something notable. https://www.amazon.com/Window-Septembers-1995-2005-Reiner-Le...

What a perfect example... I need to do this one day! Thanks for the link

For anyone wondering what he caught: "The resulting bulk of around 2,200 photographs documents both the every-day life and change of Manhattan, including the historic event of the destruction of the World Trade Center on September 11, 2001"

-- https://en.wikipedia.org/wiki/Reiner_Leist

Not a civ engineer... But the way things cascaded and experienced a domino effect seems troubling. Contrast with the ‘89 SF quake caused one upper deck section of the then cantilevered section of the SF-OAK bridge to collapse. It fell into the lower deck, but did not cause the lower deck to give way. This seems to indicate they had better tolerances (higher safery factor). But that may be a very naive take.

The article mentions this. Common engineering practice today is to avoid any single points of failure, but back when this was built, the profession was high on precision engineering based off modern finite element models and other ways to more exactly determine the capacity of a bridge.

It was deliberately built without redundancy, because they thought they could achieve perfection. In fact the models they used were correct -- it was the underlying assumptions about the long-term behaviour of the materials used that were wrong.

This isn't the first such collapse, and it likely won't be the last, but it's safe to say that engineers have learned from their failures. Modern standards generally maximize redundancy. Precision is used to reliably achieve a high margin of error, rather than to get away with the minimum possible.

So in some respect older bridges were overengineered to compensate for the lack of good modeling. Then good modelling came along and gave engineers overconfidence, now this is being corrected with better understading of materials science?

And thanks for pointing out that reliance on precision engineering might have led to safety factor reduction due perhaps to overconfidence...

Better understanding of materials, and better recognition that “once in a century” events do happen.

The once in a century thing is much less troubling than say one failure cascading into multiple catastrophic failures.

They didn’t have the capability of the time to do cascading failure analysis of that type.

Once-in-a-century actually means 1% chance per year. If you build thousands of bridges, you need a lot better than once-in-a-century, or you will have adverse events all the time.

Isn’t that affected by age and maintenance, rather than randomness? For sure bridges have finite useful lifespans, so we know in advance they need to be rebuilt or retired.

Earthquakes and other natural disasters follow a random distribution over time.

No. You can (in theory) design a bridge so that it is strong enough to withstand a 1000 year event and has a life expectancy or a single day.

An engineer may say things like “if you build it this way, do this maintenance, it will withstand 100 year events for its service life of 50 years”

You see that (in a slightly different form) with storage media. A SSD drive with a mean time between failures of a million hours typically will have an life expectancy that is significantly shorter (https://www.controleng.com/single-article/learn-or-review-th...)

I would think that depends on what kind of once-in-a-century events you’re talking about. If it is a big earthquake, chances are that, if one strikes one bridge, several others will be hit on the same day. You could have adverse events ‘only’ about every 50 years, but lose half your bridges on such days.

I don't think an assumption of uniform random distribution is fair.

Modern calculations are usually multiplied with several 'security factors'[0], anywhere between 1,1 and 1,5 as far as I can remember now (at least in Germany through a DIN [norm], and now Europe-wide via EN). These design factors come in at many different stages of the calculations, from materials to static load distribution (not sure if these are the correct terms in English)

[0] https://en.wikipedia.org/wiki/Factor_of_safety

Well, it is not like "earlier calculations" didn't take into account such safety factors, but the debate is not about existance of safety factors or their amplitude (1.3 to 1.5 has been used as far as I can remember in concrete constructions) but rather on redundancy.

To give a different example elevators/lift (those using rope cables) have usually cables that have a factor of safety 5, but they have nonetherless additionally at least one set of independent brakes.

Back to bridges and more generally reinforced/pre-stressed concrete structures, in my experience modern methods of calculation are more precise than old ones, and allow usually - given the same loads/hypothesis - to save (i.e. there is less rebar steel and cable steel) between 5% and 10% steel and/or concrete.

In practice with old methods of calculation there was a 1.05/1.10 "implied" and "hidden" additional safety factor.

The Brooklyn Bridge is the ultimate example of over-engineering to compensate for lack of good theory/modeling. It is both a cable-stayed bridge and a suspension bridge, each system of which is independently capable of carrying the entire load of the bridge. And it's also a truss bridge, though that system alone won't hold up its entire deck.

This is not so much a matter of safety factor than redundancy.

Safety factor = "it only fails if the bridge is full of heavy trucks and they all are overloaded 2X"

Redundancy = "we have 6 stays and any single one can snap and the bridge will stay intact"

Airplane wings cant have either. Because safety factor higher than 1,2 would make the plane too heavy to fly. And redundancy from multiple wings would make the system aerodynamically so unstable that snapping a single wing would still cause a crash.

In those cases you need to have very good testing program before installing the structure. And then you need a good inspection program to protect against fatigue and corrosion. The latter was missing in this case.

As a general engineering principle, it's probably a bad plan if the first indication that you've gotten something wrong is a catastrophic failure.

We still have a lot of prestressed concrete structures that predate good understanding of creep in such structures. Box-beam highway bridges, for example, were universally built from prestressed concrete after WW2, but creep wasn't well characterized until the late 1960s.

It was deliberately built without redundancy, because they thought they could achieve perfection

For some reason this reminds me of the Twitter conversation when it was discovered that T-Mobile Austria was storing passwords in plain text: "What if you get pwned?" "What if this doesn't happen because our security is amazingly good?"

Yeah, right up to the point that it isn't "amazingly good". Or one of the stays on your bridge suffers a catastrophe.

The book Why Buildings Fall Down is full of significant historical collapses and pretty much every one boils down to a lack of redundancy, a single point of failure.

http://books.wwnorton.com/books/Why-Buildings-Fall-Down/

Lessons for software engineering right there.

What is the lesson?

If something even slightly goes wrong with my software I don't want it to survive at all. This is the equivalent of putting dynamite on all the load bearing points of a structure, rigged to explode if they move out of place!

Software doesn't bend, it's digital.

That is something that is very context dependent. It is fine if it is essentially 'free' and of low criticality. But if you are dealing with an application where doing so would prove very expensive or cost lives it is clear that it isn't the ideal.

I'd say it's fine in the specific range where malfunctions can be expensive, but total failures aren't.

You pointed at one end of that, where the failure is equally expensive. From the other end of it, I attempt to build my Minecraft mod such that partial failures are transient and local, rather than bringing the entire game down.

It was deliberately built without redundancy, because they thought they could achieve perfection.

This is eerily close the the attitude of a lot of programmers.

Generally programmers try to avoid redundancy.

I guess you could talk in abstract terms about overbuilding a program by including a large amount of self-checking and possibly even error correction, but that's a bit of a stretch.

Generally programmers try to avoid redundancy.

There is the whole "Don't Repeat Yourself" mantra.

I guess you could talk in abstract terms about overbuilding a program by including a large amount of self-checking and possibly even error correction, but that's a bit of a stretch.

That's a "bit of a stretch!?" That's how you should build anything that needs to be fault tolerant and robust. Maybe not a "large" amount, but enough, such that one error doesn't strand the user in a highly inconvenient situation. A key example: text messages that can crash Messages and leave the iPhone in a state where the user has to reboot. This has happened several times over the years! Apple programmers here on HN will shrug their shoulders and tell you that you have to let the process crash to ensure memory isn't corrupted. So why not have one process which can crash, working off a file acting as a queue, with another process monitoring attempts to process each element, and limiting the number of tries for each item on the queue? Apparently, that's too much trouble, and it's easier to just pooh-pooh people on social media.

Is it "redundant" to overbuild that part of the system? Yes. It's also much better for the user and for the company long term.

You should at least build something with the assumption that it will break, and be able to reset and normal itself without loss.

A good watchdog can be more valuable than bug free software.

You should at least build something with the assumption that it will break, and be able to reset and normal itself without loss.

Too much to ask of this generation of young programmers for something like messages on the iPhone. That's not a central feature of a smartphone anyways, right? Just let the user reboot.

I meant it was a bit of a stretch to call error checking redundancy.

It's not like redundancy we are talking about in a bridge means having superfluous pillars but that it can fail gracefully that can be handled and fixed without catastrophe. Error checking can achieve a similar goal in software.

That's exactly what we are talking about with bridge design: Overbuilding it so you can lose an element without the bridge collapsing.

There really isn't a concept of a bridge "failing gracefully". Any failure is a disaster. You can talk about a bridge being functional even after it becomes weakened by the elements or overloading or poor maintenance or some other condition, but that is another case of overbuilding.

I guess what I'm saying is that software and bridges are not very similar.

Real redundancy in software development would be to do something like having three independent teams implement each module, and then use voting logic to compare the outputs. Obviously that's not practical in most real world cases due to labor costs, limited hardware resources, and excess power consumption.

> the profession was high on precision engineering based off modern finite element models

Two years ago, I organised a party for the 50th anniversary of one of Australia's first geodesic domes, which was designed and built by some undergrads in my university climbing club. It's a "hut" in a national park, and the rangers were skeptical about the structure. A 2nd year engineering student was in charge of the design: the story goes that he walked over to the parks office, showed off some printouts, and explained that the structural calculations had been done by computer. When the parks service heard that, they were sure that the structure had to be sound.

This part of the article discusses that:

> But engineers gradually recognized that the structure had so few crucial supports that if even one of them failed, an entire section could collapse.

> “There is no robustness, or the possibility of redistribution of the forces,” said Massimo Majowiecki, an architect and engineer in Bologna, northern Italy.

> That lack of redundancy, as it is now often called, “is not necessarily inconsistent with how bridges were designed in the 1960s,” said Donald Dusenberry, a structural engineer with Simpson Gumpertz & Heger in Boston.

The design was such that as long as all of the parts worked, it remained standing. But it was also such that a very small number of failures would push the remaining supports beyond their capacity to withstand the loads, leading to collapse.

The SF-OAK bridge likely had a much higher level of what this article calls "redundancy" or more generally of simply "over design" (design for loads some multiple of normal loading) such that collapse of the top deck did not push the lower deck beyond its load holding capacity. I'm surprised that "bridge design in the '60s" (as stated in the article) did not take these details into account.

The old eastern span of the SF-Oakland bridge was a series of double-decked trusses between towers. Essentially it was several loosely-coupled, independent failure domains. In failure analysis that's not really the same thing as redundancy.

the (old) Bay Bridge serviced trains until the 1960s. perhaps the design had higher loads & structural requirements due to that use.

Given that the collapse happened during a thunder storm, and given that there are reports of a lightning strike prior to the collapse [1]: how likely is it that water accumulated inside the prestressed concrete hull was suddenly evaporated by a lightning strike, effectively blowing the cable stay up? I remember reading about this theory in some interview, but never heard anything about it since.

[1] https://www.dw.com/en/speculation-mounts-over-genoa-bridge-c...

IMHO not more or less unlikely, simply impossible.

I am no expert in this field, so: why are you so sure this is impossible? If I understand it correctly, a lightning strike to the (already slightly damaged) cable stays will seek its way through the steel cables, heating them up. If there is water between the cables and the concrete, a small steam explosion will occur. If the cables are already corroded (which they were, according to the article) and thus brittle, this may cause heavy damage to them.

As the mentioned corrosion is indicative of water present, and since the cable stays are a single point of failure (also explained in the article), I personally find this a quite likely chain of events.

Presumably the bridge had a lightning protection system installed.

Check a representation of one of the stays, here:

https://static01.nyt.com/newsgraphics/2018/08/27/italy-bridg...

Basically the concrete around the steel cables is nothing but a "protection box".

From the little I know about explosions, the idea is that there is a sudden expansion from the inside to the outside. For having any damage to parts near the "core" there must be sufficient "containment", that the concrete simply cannot represent IMHO.

The actual steel used for cables is high tensile and - believe me - extremely resistant besides rather hard while concrete (particularly concrete so ammalorated to let the water creep in) has very little tensile strength.

So, IF actually there was a steam explosion, the result/effect would have been much more likely (besides "nothing") unscathed cables and ruined concrete around them (with no structural failure of the bridge)

As an alternative to explosives it is common to use expanding cement grout in actual concrete demolition, which essentially is explosive on slow motion. When dealing with reinforced concrete you normally need to cut the rebars with a torch (or other means) after the concrete has cracked.

Check (just an example this kind of cement originated in Japan in the '70's and now there are more similar products than stars in the sky):

https://www.youtube.com/watch?v=7cvcnkzK8Jk

Lightning Strike Energy == 1,000,000,000 Joules

1 Joule heats 1g water 0.24 C

1 L water == 100g

1 cu.meter water = 1000L

Room Temp == 23 C

Thus:

(77/0.24) = 320.8 Joules raises 1g water from 23 C to 100 C (boiling).

(1E9 Joules / 320.8 Joules) lightning strike with no losses along the way heats 3,116,883g of water.

3,116,883g of water == 31,168 L of water == 31 cu.meters of water. This is a block of water 1m x 3m x 10m.

Steam volume is 1600x liquid water volume. So even if we say that 90% of the energy is lost in the transfer, that's still 3 cubic meters of water that flash-boils into steam, quite possibly causing damage to the bridge.

So: perhaps unlikely, but quite possible.

Edit: 3,116,883g of water == 3,117 L of water == 3 cu.meters of water. Damn math errors, always creeping in. Still unlikely, still possible.

Indeed. Lightning does this to trees all the time, there's a lot of crazy videos on YouTube of lightning causing trees to explode.

https://en.wikipedia.org/wiki/Exploding_tree#Lightning

Trees have relatively high resistance so they absorb a lot of energy. The lower the resistance, the lower the impact.

3,116,883g water is 3,117 liters and not 31,168 liters: http://m.wolframalpha.com/input/?i=3%2C116%2C883g+water+in+l...

1L of water is 1000g, then you need additional 2261 joules per gram to evaporate. So 1E9j will evaporate 384 liters of water, so 0.384 cubic meters (unless I lost a degree of magnitude here or there myself :)).

https://www.smbc-comics.com/comics/20130616.png

Yes but the steel cable ought to conduct the electricity through the water column into the structure.

If there is sufficient resistor in the structure for massive steam explosion, then it's very unlikely that the lightning would ever strike there.

This would have been perfect situation for some destructive testing.

During night time block all traffic. Then drive radio controlled, overloaded trucks over the bridge every six months.

If the bridge fails, you just saved human lives. If it doesn't, the cost of that project is relatively small compared to pre-emptive repairs. The likelihood of regular load destroying the bridge in the meantime could be managed to be negligible.

It's already used with pressure vessels, because hydraulic testing of them is very safe. Now you would no longer need suicidal truck drivers to do it on bridges.

You forget that there are houses below the bridge, that could have been crushed. Not exactly a place for destructive testing.

If the bridge is fine, there is no problem.

If the bridge is rotten, you have lot fewer victims.

You want to do pre-emptive repairs? OK, the bridge ought to be fine. We can then proceed with the destructive testing.

You could do the testing with something like 110% of the typical max traffic load. A rotten bridge would then collapse within 6 months of it's "natural collapse". Only things you do is to make the collapse more predictable and you save lives.

I don't think a testing method where there is a risk of a lot of innocent people dying is going to be accepted anywhere. I guess that is the dilemma with these situations.

Would not be the first time saving lives ends up being less important than avoiding responsibility. So I agree on your prediction.

Besides houses below, the bridge could potentially fail after the test when in use. Better to drive these trucks in simulation before you even make your bridge.

And the Interior Minister says that the EU is to blame because it doesn't allow them to invest in infrastructure.

When the EU encourages investments in infrastructure.

Go figure.

The EU is clearly valuable to many of its partners as a domestic scapegoat. The problems come when people actually believe them and try to 'solve' the problems 'caused' by them and find a stack of disasters that makes being the prime minister a position competed /against/ becoming instead of /for/.

In the last drone video at the end of the article, the sun is quite low in the sky, making the shadows of people and tractors stretch out and play against the shapes. A horrible scene, but the scale and depth and shadows are visually stunning!

For the prudent, form follows function. Doing things because you can, because it's "neat" is courting folly. There is an analogy here for coding.

Are there any experts here who can elaborate on whether Professor Gentile's acoustic testing of the bridge's cable stays may have had any negative effect on its stability?

I found this paper on the method: https://www.ndt.net/article/jae/papers/20-083.pdf

E: as mentioned below, the method is completely passive.

You may want to get in touch with [Professor Raimondo Betti][1], who [I spoke][2] with at Columbia University prior to a talk he gave about his research on [a bridge cable corrosion monitoring system][3].

He was optimistic about the ability of his (from my understanding, passive) sensor system in helping prevent these tragedies, but pessimistic about the political will to apply it to bridges.

[1]: http://civil.columbia.edu/raimondo-betti

[2]: https://www.simongriffee.com/notebook/GRS-20151015-100821/

[3]: https://www.youtube.com/watch?v=cP1CfGVVXO4

Many thanks! Just out of curiosity: why was he pessimistic about the political will to apply this method? Because politicians fear the results?

I am not sure, but that sounds right.

This was a few years ago and my memory may be faulty, but I remember him saying that it would probably take the collapse of a large bridge like the Brooklyn or George Washington bridges in NYC before the authorities in the city would consider using the system.

From a political standpoint you are spending money that might tell you that you need to blow your budget on rebuilding some infrastructure right now. Bridge collapses are so rare that nobody thinks it will happen to them. Just ask the people of Minneapolis.

Be super careful with that line of thought: it is precisely what they will use to deflect blame from the construction, engineering, and maintenance issues. "The bridge wouldn't have failed if it weren't for that pesky diagnostic procedure that also happened to raise all sorts of problems about corrosion and structural integrity"

edit: also - not an expert but interested and somewhat informed - this is a very common structural engineering technique and should under no circumstances, ever, lead to this kind of outcome on a sound structure.

Also Italy has a (recent) history of prosecuting scientists who attempt to give their opinion (even if it isn't listened to). If I was an Italian scientist I would stay far away from predicting anything to or for the government.

http://www.sciencemag.org/news/2015/02/why-italian-earthquak...

Reading that paper, it seems it solely comprises attaching sensors (I'd imagine piezo transducers bolted firmly onto measurement points, most likely) and listening - so it's hard to see how it could have had any effect, unless they attached the sensors with a method that would damage the bridge, which seems unlikely.

Yes, I also just (slowly) realized this while skimming over the paper. So it is indeed completely passive.

What I find unusual is - according to the infographic - both southern cables broke in the initial stage. And then that was followed by the northern cables breaking in response to the load shifting.

I presume the two southern cables are independent components and support independent platforms. Why would they both break at the same time, if not for some exogenous event?

It’s amazing to me in this day in age that there are no actual videos of the collapse. Not even from a car camera. I’ve searched everywhere and it would really be a very useful piece of information to have in order to diagnose what really happened.

Per the article, the collapse was recorded by a security camera and the footage is being analyzed by the investigators, but it has not been made public yet.

Use stainless steel cable instead?

It cost ~4-5x times more than regular (today, not sure decades ago) but total cost of bridge build for it's majority was probably (overpriced) labor anyways.

In machinery stainless steels have universally poorer mechanical properties vs non-stainless alloys - a reduced corrosion rate is about the only benefit, and even so, it's still just "stain-less" — not "stain-free" — over the timescales of civil infrastructure, it would likely suffer damaging corrosion at some point.

The price ratio is roughly 6 historically, it has lowered to as low as 4 only recently, and stainless steel has been largely used, since the '80's, instead of conventional rebar steel.

But there is no stainless steel capable of having the same resistance/elasticity of cable steel.

To give you some comparative data in EU the normal rebar since more than 40 years is usually grade 44 or 45 that means that it starts elongating (and eventually fail) around 4300-4500 Kg/cm2 and breaks over 5400 Kg/cm2 (still as a reference normally the calculation uses a max of 2600 Kg/cm2).

Cable steel (again since more than 40 years), typically in 7 wire strands, is almost 4 times stronger, the same values are 16700 kg/cm2 and 18600 kg/cm2.

Typically it is pre-stressed to 12500-13600 kg/cm2.

Stainless steel cable is used in the fixed rigging (mast stays) of yachts precisely because of it's anti-rusting properties. Unfortunately the stainless still has to be completely replaced every 5-10 years because it weakens to the point you can't insure the yacht anymore.

Stainless has a very different failure mode to normal steel where a tiny nick in the surface of the stainless under tension forms a crack and corrosion occurs in the very bottom of the crack making it propagate deeper into the stainless until it fails while still looking shiny on the outside.

Reply on news.ycombinator.com