Hacker News new | past | comments | ask | show | jobs | submit login

https://en.wikipedia.org/wiki/Tretten_Bridge

The ironic part is "The bridge replaced a steel truss bridge built in 1895 at the same location". Engineers in 1895 knew how to use steel trusses and built a bridge that lasted 100+ years, and in 2012 they decided to use glued wooden beams and it fell apart in 10 years.

Well perhaps Plan Arkitekter and Norconsult can take a trip to the archives and see what the engineers in 1895 knew before building any more structures.




> Engineers in 1895 knew how to use steel trusses and built a bridge that lasted 100+ years

It's a little too strong of a claim. Plenty of poorly designed steel bridges from the 19th century have collapsed as well, despite the superior material.


> Plenty of poorly designed steel bridges from the 19th century have collapsed as well

That's true. But I only meant that particular bridge, which did last for 100+ years. So, when the architects and the engineers go to the archives to relearn a few things, they can focus on the ones which haven't collapsed.


To be fair, there have been a lot of advancements in timber and glue -- They just didn't take into account their weaknesses when designing (e.g. Stress that would pull laminated beams apart).


There are always unknown unknowns in trying new designs and/or materials. Even with all the simulations and calculations, nothing beats experience.

Sure, novel things should be tried, but in a conservative manner. Overdesign first, then optimize iteratively. Because failure costs lives.


I believe the failure happened in the joint between the wood and steel. The joint was a sandwich with very thin slices and with many nails. Anyone with materials intuition knows that puncturing a material to such a degree weakens it. Mix in constant moisture and single points of failure.. Should have been a couple of big old bolts and some bracings, alternately thicker slices in the sandwich


Glulam isn’t a new material, it’s over 100 years old.


How many 70 m span glulam road bridges have been built in those 100 years?


Yeah, exposing load bearing wood and glue to the elements requires some extra caution and engineering intuition


Isn't that basically arguing "Don't talk to me about statistics, I want to talk about this one specific data point"?

Bridges on the whole are pretty clearly better built and safer today then they were 130 years ago. I don't think there's any significant argument to the contrary. There are outliers in both directions, obviously. The old one at this site looks like it was pretty great. The new one sucked. That doesn't say much about why that was true, just that it was.


On the whole they may be better built, but the same design in steel would likely not have failed?

I guess that’s what they’re going for when they say aesthetics had too much of a focus?


> On the whole they may be better built, but the same design in steel would likely not have failed?

I don't see how that follows? Lots of steel bridges have failed over the past 130 years. The overwhelming majority of new-built bridges or any material don't fail.

People really, really want this to be about "aesthetics" because it confirms their priors about being smarter than the rest of the world or whatever. I'm just saying that's 100% bunk and that there's zero evidence for that hypothesis.

Architecture has had a "strong focus on aesthetics" since the very first stones were piled up to make a temple. Let's not get ahead of ourselves here.




> Isn't that basically arguing "Don't talk to me about statistics, I want to talk about this one specific data point"?

For sure, we can each talk about different things. I can talk about ironic things and you can talk about statistics, and we can still be friends!

> That doesn't say much about why that was true, just that it was.

Well, why do you think it sucked?


There's nothing in statistics that helps you build a bridge.


> Isn't that basically arguing "Don't talk to me about statistics, I want to talk about this one specific data point"?

And why not? In this particular case, the statistics did not justify the expectation that this bridge would be at least as robust as the one it replaced.

There's a general point to be made about engineering progress, but here we have a case which bucks the trend. Speaking in generalities when the topic is a specific incident is a way of missing the point.


+ why even remove the steel bridge in the first place?


Moisture causes corrosion.

Rivers produce lots of moisture.

Corrosion weakens the steel. It's literally the metal flaking away.

All steel corrodes. Even stainless.

Once corrosion sets in and the outside of the metal starts to pit, the corrosion speeds up drastically. (The old area vs volume thing, same thing that makes lump coal difficult to ignite while aerosolized coal dust is practically an explosive).

Here's a modern US bridge - all steel - that collapsed in Pittsburgh two years ago. In some areas over 80% of the original steel thinkness was gone after 49 years.

https://www.youtube.com/watch?v=J-VnWB4fiFk


Paint protects steel. A frequently painted steel bridge can last centuries.

As always, lack of maintenance and poor design choices (use of corten steel) are the culprit.


Ode to The Paint on Nuclear Submarines

    The purpose of the hull is to protect the paint.
    The purpose of the reactor is to drive the paint around.
    The purpose of the SUBSAFE program is to ensure the paint comes to the surface and will not be lost.
    The purpose of the cathodic protection system is to back up the paint.
    The purpose of the weapons is to defend the paint.
   The purpose of the Special Hull Treatment is to protect the paint.
   The purpose of the Vertical Launch System is to destroy those who would do the paint harm.
--Naval Research Lab


What's the purpose of the crew? Provide the paint company?


The purpose of the crew is to reapply the paint.


They are the only carbon-based life forms, otherwise useless to the paint, capable of touching up the paint.


The Golden Gate in S.F. has been continuously painted since it was built in 1937 ... that when they've finished on one end, they move to the other end and start again. [0] https://www.goldengate.org/bridge/bridge-maintenance/paintin...

This page details the other continuous maintenance going-on. [1]https://www.goldengate.org/bridge/bridge-maintenance/


From reading the linked article it seems that they are continuously painting but it’s not really end-to-end then restart. It’s more of an ongoing maintenance as-needed, at least based on the description on that website.

Now that effectively might mean then combing over the bridge end-to-end and painting as needed so what I said might be a distinction without a difference.


There so far haven't been any steel bridges that have actually lasted centuries, even if we highly suspect that they are capable of it with the right maintenance.


Good point. The really old ones are iron.

ETA: Also, the ones that have survived aren't road bridges. They're either railway bridges (and thus were built strong in the first place) or have been relegated to historic preservation/pedestrian use only. Modern cars and (especially commercial) trucks are far heavier than early vehicles. A road bridge built in the 1850s was only designed to carry foot traffic and horsedrawn vehicles.


Something like 20% of the bridges in this country are 10 years or more past their design lifetime and mostly not due to be replaced any time soon. You think they're keeping on top of painting every one? Saying preventing corrosion is as easy as slapping on paint is like saying computer security is as easy as not leaking your password. There are a million things that could happen, many of which are outside what any single person could control.


I don't know in what poor country you live, but in my country bridges are well maintained and I don't recall a single critical failure. Even though we're one of the poorest EU countries.


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

"There are more than 617,000 bridges across the United States. Currently, 42% of all bridges are at least 50 years old, and 46,154, or 7.5% of the nation’s bridges, are considered structurally deficient, meaning they are in “poor” condition. Unfortunately, 178 million trips are taken across these structurally deficient bridges every day. In recent years, though, as the average age of America’s bridges increases to 44 years, the number of structurally deficient bridges has continued to decline; however, the rate of improvements has slowed. A recent estimate for the nation’s backlog of bridge repair needs is $125 billion. We need to increase spending on bridge rehabilitation from $14.4 billion annually to $22.7 billion annually, or by 58%, if we are to improve the condition. At the current rate of investment, it will take until 2071 to make all of the repairs that are currently necessary, and the additional deterioration over the next 50 years will become overwhelming."


I don't know in what rich poor EU country you live, but in my "rich country" bridges are basically left to their own devices until a critical problem occurs and then they hastily start repairing it. In some cases they've been repairing bridges for decades that way. Some have recently finally actually been replaced or are being replaced. But nobody is going out and replacing a bridge that isn't on the brink of collapse and held up by "essentially duct tape".


Now do a quick internet search for “hammersmith bridge”.


Which is a shame as it is otherwise a beautiful bridge.


maybe they wouldn't be so poor if they didn't waste all their money on infrastructure maintenance


What else is a country supposed to spend money on‽ Transportation, utilities, healthcare, bureaucracy, government: a country is its infrastructure.


There are metal bridges over 100 years old still in service today, including a few you've heard of. Do you know this specific bridge was replaced for this specific reason, or are you just making stuff up?


I’m not making any thing up. Why the hostile tone?

This is basic materials engineering.


The question was why this specific steel bridge was replaced. Not "what are possible reasons someone could replace steel bridges?" and it's absolutely not "basic materials engineering" that every bridge needs replacing after a fixed amount of time due to corrosion.

So again, do you know this specific bridge was replaced for this specific reason, or are you just making stuff up? Because bumping in on a specific question with a generic answer without any knowledge if that applies to this specific case is making stuff up.

The hostile tone is because you're talking complete nonsense with great arrogance.


You have more pictures on the Norwegian Wikipedia page: https://no.m.wikipedia.org/wiki/Tretten_bru#/media/Fil%3ATre...

The old one wasn’t wide while the new one had two lanes and a normal sized sidewalk.


> Plenty of poorly designed steel bridges from the 19th century have collapsed as well, despite the superior material.

Disagree. In late 19th century steel girder bridges were well understood and many of them survive to this day.

They were usually replaced due to changed demands (wider roads, heavier loads), not because of failure.


Is this late enough in the 19th century for you? https://en.wikipedia.org/wiki/Tay_Bridge_disaster


So your convincing evidence for the claim that "plenty of poorly designed steel bridges from the 19th century have collapsed" is a single example of a bridge that collapsed because it was designed "less extensive and robust than [..] previous similar designs"?

Or is your claim a mere circular argument, since we can deduce that a steel bridge was poorly designed if it has since collapsed?



“Plenty”?


Well to paraphrase a common saying, it's easy to build a bridge that stands, but it's difficult to make sure that is so optimized that it's on the verge of failing but still somehow stands.


A good engineer can make a bridge that stands 1,000 years. A great one makes one that can stand 100 years.


These guys must be the best engineers in the world then!


Well I... umm... hmmmm...


The bridge was built during a trend to build wooden apartment houses here in Sweden, and probably all over Scandinavia.

This fad has now passed and there have been several news stories about what a bad idea it was. I guess they were trying to get back to using more natural materials.


There were 250% more apartments built from wood built in 2022 than in 2012. In total 14.4% of all new apartments finished in 2022 used wood.

https://www.di.se/hallbart-naringsliv/trahustrenden-nar-nya-...

https://www.tmf.se/bransch-naringspolitik/branschutveckling/...


That's true, I had missed how popular they had become. My brother briefly lived in a wooden apartment building around 2017 and after they moved out I noticed a few criticism of the new wooden trend. Then I just lost track of it.

The trend goes back to 2014 even, so it wouldn't surprise me if this wooden bridge built in 2012 was part of that whole wave.


There's nothing wrong with wood frame construction, if it's done right.

It sounds like they didn't do it right, though.


Am I the only one who thinks it's strange there isn't a single picture of the bridge pre-collapse on that page?


there is one though!


You will see wooden covered bridges from 1895 that last a long time too. It is not a material choice problem here, it’s the design and how the members are laid out.


Survivor bias.


But just think how beautiful it is.

Well, was.

Sarcasm aside, I see more beauty in doing something well than whatever this was. Are they hiring art students to build bridges now?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: