Hacker News Reader

Comment by mabbo | Hacker News Reader

There are always ways to build the tall buildings safely- they're just more expensive. These buildings aren't unsafe because there's no choice, they're unsafe because the builders wanted to save money and the building code was too loose to stop them.

Japan is a highly volitile region, tectonically speaking, yet they've managed to build lots of tall buildings safely. How? By having some of the strictest building codes in the world, yet allowing development within those rules.

Meanwhile, outside of SF itself, the Bay area refuses to build any real density that's desperately needed, using earthquakes as (one of many) excuses for NIMBYism and entrenched interests of landlords.

View on HN

As an architect in San Francisco, I would definitely prefer to be in a modern high rise here over other buildings for safety purposes during an earthquake. I am highly skeptical of this article as I have in depth knowledge of seismic engineering, the building code and the updates after every earthquake around the world to the code and engineering practices.

The leaning building across from my office, for example, has nothing to do with seismic issues but basic design/engineering/construction flaws in not extending the pile footings deep enough to solid ground as others in the area have such as the new Salesforce tower.

If the earthquake were large enough, could the Millenium Tower completely fall over? Or is it only destined to keep leaning?

High rises in general are designed to sway in an earthquake - that sway absorbs the horizontal forces being applied - similar to a tree. If an earthquake gets sufficiently large, say larger than any known measured earthquake to date, that sway differential would increase enough that buildings would either hit each other, or the building would sway beyond the ability of the structure to recover and fall over. A standard approach for tall buildings is to incorporate either a dynamic weight high in the tower to counterbalance the sway so it doesn't go critical, or incorporate dynamic elements in the foundation system. In general, if a modern high rise is falling over, all buildings without the benefit of this level of modern engineering have already fallen over.

Would you be able to provide an explanation of what a dynamic element in the foundation is, for example?

See how a dampened dynamic build vs a regular dynamic build react to movement (and you can imagine the static build): https://youtube.com/watch?v=xp2pGxFzrzI

What is your background?

>Apologies for the complaint, but could someone please let me know why I'm being downvoted for this question?

It's likely because you're implying "dynamic" is a "buzzword" in a discussion about structural engineering and seismic movements (which is, by definition, a dynamic system).

But I'm not, I was saying that due to my background, I can't help but think dynamic is a buzzword. Therefor I was asking someone who was clearly knowledgeable on the subject to set me straight.

I don't think linking it to a buzzword adds anything to the conversation. The question didn't need a justification.

All right. I've removed it. Such is life

The "why"s (your prejudices) are irrelevant.

Also I don't really think there's a background you can have where you don't understand, or cannot lookup, the difference between dynamic and static if you really cared to understand. Seems more like self-important rhetoric than anything.

In mechanical fields, dynamic means "moving". The weight literally moves, so it's dynamic. Another YouTube video that explains it a bit better: https://www.youtube.com/watch?v=f1U4SAgy60c

> A standard approach for tall buildings is to incorporate either a dynamic weight high in the tower to counterbalance the sway so it doesn't go critical, or incorporate dynamic elements in the foundation system.

As a structural engineer on the west coast, I can say that neither of these are standard. Tuned mass dampers (TMDs) are commonly used for reducing wind-induced vibration, but not for seismic applications. Nor have I ever designed, seen, or heard of 'dynamic elements in the foundation'

The reason you can't rely on TMDs is because they are tuned to a structure's elastic period of vibration. That is to say, the materials are still in their elastic range. When a ductile structure is subjected to a sufficiently large earthquake, elements of the seismic force resisting system (SFRS) will yield. Hence the period will elongate. Hence the damper will be detuned and may not provide any benefit at all.

I think he is referring to base isolation which, I agree, is non-standard.

I meant standard for the really big buildings under the highest stress. For lesser challenges, agreed, non-standard.

They don't base isolate tall buildings, they isolate short ones. The whole point of base isolation is period elongation. In a tall building, you're already flexible, and your column loads are tremendous. It makes no sense.

A link to the next generation TMD's to handle earthquakes: tuned mass blocks, orthogonal poles, and torsional pendulums (TMDPP) : https://www.hindawi.com/journals/sv/2017/5834760/

1) I'm not saying there aren't novel solutions - there's a whole category of devices that use active control. I'm saying its not common place.

2) From what I understand, China has a completely different design philosophy. They design structures to remain elastic, under certain earthquakes, which would then allow for TMDs.

Are design parameters such as pile depth part of the public record? Is there a way to lookup the design of a building?

Most cities keep a set of permit drawings, but generally you need to have a proximate relationship to the building to get access to them

> The leaning building across from my office, for example, has nothing to do with seismic issues

Nowhere in the article did anybody make such a claim. It was to illustrate the general problems of building in that area even without an earth quake, which leads me to the other point, what IMO is the main issue, and I think is pretty independent of how much effort is spent on any one building: The potential for soil liquefaction. They included several maps showing that many of the tallest buildings are built pretty much exactly over ground that might suffer that fate in a big earth quake. How would any individual design help? Compared to the alternative of not building the tallest structures right over those spots. The maps also showed plenty of more solid ground, but that is where there are mostly lower buildings.

I would like to hear your opinion on that issue.

As for your doubts on the entire article, it seems to me they asked quite a few specialists? Are they all mistaken?

Well, I certainly would like you to be right, I used to live in SF and I still love the entire greater Bay Area.

Liquefaction mitigation is possible (ground replacement cells, compaction grouting, etc). It costs a lot and the only place that we (my company) have done true liquefaction mitigation is on large state funded jobs. The preferred alternative for the Bay Area is to sink foundations through liquefiable soils, discounting any strength from these soils, and rely on empirical studies that have found that liquefaction does not necessary result in loss of confinement which would induce buckling failures in the piles.

I share the GP's doubts on the article. The practicing specialists they talked to certainly don't think there's a gamble: they actively pursue and design sky scrapers in the City. I think their words are being misused in a way that makes them seem like they agree with the article's premise.

Your comment seems to echo that of another architect quoted in the article.

Ron Klemencic, the chief executive of Magnusson Klemencic Associates, the company that did the structural engineering for Salesforce Tower, says he agrees that water and sewage systems need higher strength requirements, but not high rises. “Buildings falling on top of other buildings — that’s not going to happen,” Mr. Klemencic said.

They never explained why Klemencic's reasoning. Care to explain why "buildings falling on top of other buildings — that’s not going to happen"?

My understanding is that the way buildings collapse in Hollywood is not how they collapse in reality. If the base of the structure begins to fail, it doesn't fall over as a tower, but crumbles into itself.

The Twin Towers took out the blocks surrounding them. Literally they fell on top of other buildings, even though they "collapsed into themselves."

I am very skeptical of anybody, no matter how experienced, that gets on the news and says "that'll never happen," regardless of what the "that" is.

Has he considered a skilled terrorist cell lacing only the east side of Salesforce Tower with explosives?

Even blowing out only one side of a tall tower, would not cause it to fall like timber. Instead, one of two things will happen. If there is low lateral strength (unlikely), the east side of the floors above will shear and collapse off, straight down into a pile on the east side. If there's good lateral strength, The forces formerly supported by the now blown east structures will largely fall on the adjacent structures around the building, overloading them, causing failure, and rapidly continuing around the supports in a zipper-failure, whereupon the above structures now unsupported again fall straight down with gravity's vector.

Either way there's probably a bigger pile on the east side, but it's not like the top floor of a 1000' building will land 1000' laterally from the base. 100' would be a long way (tho there's be lots of flying debris for a good distance).

Skyscrapers have a tremendous amount of mass and are engineered to resist the force of gravity. That's hard, so they only bother to engineer it so that it resists that force in one direction relative to the building.

As soon as you might start to turn the building onto its side it falls apart like a sand castle. Were this not to happen the building would have to be engineered such that it could withstand being turned on its side like that without collapsing. But they don't design them as such.

Wouldn't there be bigger concerns, terrorist attacks really don't happen that often.

It'd be like constantly worrying about protecting yourself from lightning.

You're right, but over the course of a building's lifespan, the "chance" is greater, right? So the chance of a terrorist attack against salesforce tower this year is less than the chance it might happen sometime in the next 30 years. I'm not good at statistics, though.

Anyway, this is why we have a lot of earthquake preparedness, though the chance is 5% in the next 30 years.

It's not a "constant worry," so much as a plan. There's non-negligible chance of terrorist attack, so they architect in such a way to prevent it being catastrophic. There's a non-negligible chance of your office catching fire, so they put green exit signs at the door. That sort of thing.

> though the chance is 5% in the next 30 years.

Not even close. USGS estimates over the next 30 years are: 72% probability of a M6.7 or higher 51% probability of a M7.0 or higher 20% probability of a M7.5 or higher

Source: https://pubs.usgs.gov/fs/2016/3020/fs20163020.pdf

And?? California building codes have improved alot since the 1980s or so. I remember the 1989 LA earthquake as I drove through the aftermath to visit my grandma about 2 weeks after, I live on the Central coast.

All in all not since 1933 has CA lost more than 100 lives in an earthquake: https://en.m.wikipedia.org/wiki/List_of_earthquakes_in_Calif...

That tells me our building codes are pretty damn good.

I work at ucsb and the catilina islands recently had a 5.2 about 2 weeks ago and the building shook abit. But all in all unless we suffer an 8.0+ I'm not terribly worried (knock on wood). To put that another way, realistically for any moderate sized earthquake CA is generally well prepared in terms of architecture, unless the big one hits ... In which case just kiss your butt goodbye.

I do wish, even for the big one, that CA invested in an early warning system like Japan has ... Imagine a 5-50 seconds warning of an earthquake, that would save lives, not stricter building codes: https://youtu.be/OXXZouxPT7U

Here is a better, live example of the early warning: https://youtu.be/n-FMpNBfna8

An early warning system might be nice, but stricter building codes are what is going to save lives. Japan has those too.

It's also worth pointing out that there has arguably not been a large quake close to a major city center in California since 1906 (for some values of "large" and "close"). The 1994 Northridge quake, which did some $15B in damage, was probably the best recent preview we have. Loma Prieta certainly did serious damage, despite being (as I recall) some 75 miles from San Francisco.

The real test is when the Calaveras Fault that runs up the east side of the SF Bay breaks. The last time was 1868, when there wasn't much there; estimated magnitude was 6.3 to 6.7. A quake of that size on that fault now — and it's getting to be likely, in the next two or three decades — is going to make a hell of a mess. We'll find out then how well a lot of things hold up.

Earthquakes are interesting, I was mostly talking about terrorist attacks, which are basically rounding errors when compared to other causes of death.

I don't think it would cumulatively add up to much, but would love to see the math.

>It'd be like constantly worrying about protecting yourself from lightning.

A lightning rod (US, AUS) or lightning conductor (UK) is a metal rod mounted on a structure and intended to protect the structure from a lightning strike. If lightning hits the structure, it will preferentially strike the rod and be conducted to ground through a wire, instead of passing through the structure, where it could start a fire or cause electrocution. Lightning rods are also called finials, air terminals or strike termination devices.

In a lightning protection system, a lightning rod is a single component of the system. The lightning rod requires a connection to earth to perform its protective function. Lightning rods come in many different forms, including hollow, solid, pointed, rounded, flat strips or even bristle brush-like. The main attribute common to all lightning rods is that they are all made of conductive materials, such as copper and aluminum. Copper and its alloys are the most common materials used in lightning protection.[1]

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

With the Twin Towers, the issue wasn't so much that they fell on top of other buildings (they didn't - you can see that on footage of them coming down), it's that when they hit the ground, the debris cloud billows out at high velocity and consists of projectiles that can structurally damage neighboring buildings. All that gravitational potential energy of the structure gets converted to kinetic energy as it collapses, and that kinetic energy is basically like a bomb going off at ground zero. When a bomb goes off near a building, the building is gonna have a bad time, even if it doesn't directly hit it.

http://s3.amazonaws.com/nasathermalimages/public/images/4158...

http://s3.amazonaws.com/nasathermalimages/public/images/fema...

http://s3.amazonaws.com/nasathermalimages/public/images/5489...

http://s3.amazonaws.com/nasathermalimages/public/images/core...

http://s3.amazonaws.com/nasathermalimages/public/video/Relic...

http://nate.flach.s3.amazonaws.com/Abolhassan_Astaneh.avi

Like I said, it's like a bomb going off at ground zero. The pictures you link support that: notice how the damage is mostly near street level (from projectiles moving outwards from the crash site) rather than at the top of the building (from projectiles falling on the building).

I agree. https://web.archive.org/web/20021004010033/http://www.chiefe...

http://s3.amazonaws.com/nasathermalimages/public/images/Impa...

Hollywood treats falling buildings like falling trees [1]. The building basically gets cut off at the base, then stays together as it falls to the side until it hits the ground. I'm no expert, but IRL I think any significant amount of lean would lead to massive structural failure in your average skyscraper. Once the structure starts to fail gravity is going pull the pieces straight down since nothing is pushing them to the side. It's certainly a danger to surrounding buildings (see the WTC) but vastly different from buildings falling over on their sides to crush their neighbors.

1: https://youtu.be/r0tgEYiTNd4 (0:35 and 0:45)

I don't agree with Klemencic on this - given a big enough earthquake with large enough building sway (imagine a tower being a vertical beam with the base fixed and the top swaying back and forth) the structure can fail such that the momentum of the building during maximum sway will either break the building and/or cause the building to impact other swaying buildings.

Simulation showing an actual building failure from the NZ Christchurch earthquake [1]. It collapsed in the Feb 2011 Christchurch earthquake, magnitude 6.2[2].

Once the building's supports fail, it drops rather than topples.

[1] https://www.youtube.com/watch?v=TCJq_2-q34k [2] https://en.wikipedia.org/wiki/2011_Christchurch_earthquake

That's a squat cube-like 5 story building. You're probably right, but that video alone doesn't convince me a skyscraper would necessarily fail in the same way.

It wouldn't. The are designed to withstand hurricanes. It takes extensive preweakening to bring one (or three) down.

People are more likely to be killed by flying glass in the Financial District than skyscrapers collapsing. A good approximation of how far glass can be thrown downtown is roughly 2x building height due to the winds coming off the bay.

Is that supposed to make it better?

I have friends who unfortunately got to witness what happens when plate glass comes down. They can barely talk about it. It also means that the injured and trapped folks in the high rises are going to be without aid in buildings with dead mechanicals for awhile.

I think he's saying, in an earthquake, stay inside.

A lot of people think running into the street is what you wanna do. It is, in like, Mountain View maybe. Realistically you want to just jump under the nearest thing you can and hope for the best. If you're in the Financial District in the middle of the street... well your friends apparently have seen what that'll look like.

Right, that was the point I was trying to make, perhaps I didn't state it well. Seek shelter immediately and try to avoid flying glass. I would also avoid UMBs (Unreinforced Masonary Buildings) these buildings WILL collapse. They aren't as common in FiDi but are all over Soma, Chinatown, and the Tenderloin.

Yeah. Lots of stuff comes from above if you are on the street. Going inside a building is a good bet.

"35 deaths and many injuries from falling masonry during the 2011 Christchurch earthquake. Buildings with dangerous parapets, facades and verandahs" [1]. That was 20% of total deaths.

Certainly near a skyscraper, you want to go inside, since if it pancakes, being outside won't help anyway - debris will hurtle horizontally.

[1] https://www.stuff.co.nz/business/71738867/property-council-s...

> like, Mountain View maybe

During the Loma Prieta Earthquake a bunch of my coworkers fled the building and took shelter in the parking lot under the power lines.

A modern high rise- but only as long as you are confident there were not basic design/engineering/construction flaws?

Which, for those of us who are not structural engineers and cannot review the work to see that it is indeed well done...

This gets at my concern though - it is not that we don’t know how to build skyscrapers safely. It is that there is now positive evidence that those construction and engineering techniques are not being applied properly, at least in some cases.

For me, the question is not about whether high quality projects are going to be okay. The question is: on what basis do I believe there has been sufficient oversight to guarantee the soundness of new construction?

I can remember talking to SPUR members in 2001 about how problematic it was to extend downtown development into SOMA, etc, due to liquefaction and uncertainty. When the building craze hit, it felt like it brushed these concerns aside, rather than answer them. Now we have leaning buildings that don’t meet basic construction requirements. It doesn’t inspire a lot of faith.

The leaning building across from my office, for example, has nothing to do with seismic issues but basic design/engineering/construction flaws

You say that, and then you claim that seismic engineering is free from such flaws? That seems a bit like hubris.

Seismic is different than basic construction. That building tilt would have happened anywhere vs. in an earthquake prone zone like SF.

Japan and California are pretty comparable in terms of science and engineering for earthquake resistance. There are some differences on the public policy side, which influence how the engineering requirements are implemented, but Japanese buildings are not intrinsically safer. Note that the 1994 Northridge (Calif.) and 1995 Kobe (Japan) earthquakes were about the same side - the Japanese quake was ~10x as costly and ~20x as deadly (mostly due to population density around the epicenters).

Yeah. While I don’t support NIMBYism, YIMBY-types seems to love to use Japan as their counterexample, without understanding that Japanese homes tend to be built somewhat dangerously and without regard to earthquake safety.

Building standards were first implemented in Japan in 1952 and were revised in 1981 and 2000. Apologies for not having an English source, but of the people who were killed because of the house or building that they lived in, 98% were living in a building/house that did not meet the standards of the 1981 revision to the building codes.

https://www.kobe-np.co.jp/rentoku/sinsai/20/rensai/201409/00...

Isn't there is a slight difference - the maximum expected Japanese earthquake (9.0) magnitude is much more powerful than the maximum expected California earthquake along the Calaveras/San Andreas (8.0)? Cascadia looks to be another 9.0 fault line but the epicenter would be closer to Seattle/Portland than California.

How would you build to withstand against soil liquefaction? It could be likely that Tokyo is built on more stable bedrock.

https://www.sciencedirect.com/science/article/pii/S003808061...

> The 2011 Great East Japan Earthquake caused the severe liquefaction of reclaimed lands in the Tokyo Bay area, from Shinkiba in Tokyo through Urayasu, Ichikawa and Narashino Cities to Chiba City.

Those aren't the parts of Tokyo with skyscrapers.

Most of Tokyo comprises low-rise (1-10 story) buildings.

Which parts have the tallest buildings?

My understanding (probably naive) is the footings driven into the ground are deep enough to go below this area and rest on stable rock.

That’s how it was explained in a documentary about mission bay (south of AT&T).

edit: For those interested, this documentary explains the approach https://www.youtube.com/watch?v=SyjijQS2dAk&t=1219s

Do you remember the name of this documentary? A quick search didn't turn up anything. I live in the area and would love to learn more about this.

I recall seeing some computer animation to this effect in a documentary (in the UK) about the Millennium tower tilting - searching on that brought up some confirmation in the form of a press release by the TJPA:

http://tjpa.org/uploads/2016/08/301-Mission-Press-Release-Au...

The 60-story Millennium Tower is made of concrete rather than steel, resulting in a very heavy building. This heavy structure rests on layers of soft, compressible soil. The foundation of the Tower, however, consists only of a concrete slab supported by short piles that fail to reach the bedrock below. That foundation is inadequate to prevent settlement of a building with the weight of the Tower. In contrast, the Salesforce Tower and 181 Fremont Tower, also adjacent to the Transit Center, are supported on piles drilled down to bedrock. Millennium Partners’ poor design decision is the cause of the tilt and excessive vertical settlement of the Millennium Tower.

Sure, at about the 20:00 is when they discuss Mission Bay specifically. The whole documentary is solid though if you have more time. I learned a lot and found it very fascinating.

https://www.youtube.com/watch?v=SyjijQS2dAk&t=1219s

Either go deep enough to hit bedrock or design your building to float. They both work fine, it's just a question of cost.

Well, option #1 is to build somewhere else. Is every square foot of SF subject to liquifaction? Or is it just the cheapest piece of ground? Is it that the local government has zoning rules that simply prevent development and density in the places that would be safer? To be clear, I don't blame developers for finding the cheapest option- that's how capitalism works. It's up to governments to make the rules such that the cheapest options are also the best ones.

Second, is there anything that can be done such that the soil won't suffer from liquifaction or will to a lesser degree? Probably expensive to do, yes, but surely there are means.

And there's probably a million other ideas that might work that I don't even know of. It's silicon valley, that's where ideas come from, right? "Impossible" usual gets San Francisco types all excited.

There's a downvoted reply to my original comment saying that I haven't actually tried to build these- and they are correct. I'm not a civil engineer. But I am a firm believer that if you make strong regulations and enforce them, clever engineers always find a way, at a cost. But the hard part of that isn't the engineering, it's having the political willpower to enforce strong rules.

Japan is always touted for this, but Chile (Santiago, Viña) are similarly plagued with extremely large earthquakes, with regular occurrence over 7.0, and you see tons of mid-rise buildings in Santiago. The building standards there are exceptionally high after the 1985 earthquake (8.0). If Santiago and Viña del Mar can do it, surely San Francisco can.

Also, If I recall correctly, what did the most damages in the 1906 SF earthquake was not the actual earthquake itself.

What caused the most damages was the fires after the earthquake.

Hopefully SF has a better fire department now than in 1906.

same thing for Kobe in 1995. Fires were responsible for most of the deaths.

It's not just the fire department that's better - in 1906 a lot of those fires were caused when the quake would pull a gas appliance from a wall and leave a gas leak that later ignited.

Nowadays everyone has their hot water heater and oven bolted or strapped to the wall. I'm sure there are other useful fire codes California has adopted since then but this one rule will probably prevent more fires in the next big quake than any other.

Additionally, gas fittings that are designed to (safely) breakaway are more common.

My LA house has to have a shutoff that automatically kills the main gas line during an earthquake. One must be installed when the house changes hands. No grandfather clause exception allowed.

yes, i'd forgotten about those; we got one installed at our house a couple years ago, iirc at no charge from PG&E. it's definitely better/safer than the per-appliance solutions.

Going a bit deeper, the earthquake took out most of the water supply to the fire hydrants in the city, and so the fire department couldn't put out the fires that started during the quake.

As a result, SF now has a backup hydrant system! Those big fat multi-colored hydrants are all connected to a fire-only water system that's gravity-fed from cisterns all over the city. The colors of the hydrants encode information about which cistern the water comes from, IIRC. And whenever you see a random arc of a circle made of brick inlaid in the road, that's outlining the location of a cistern.

https://www.atlasobscura.com/places/the-golden-fire-hydrant https://www.citylab.com/design/2017/05/the-sublime-subterran...

I don't have the link handy, but I was reading an article about how the backup water system is inadequate for the western part of the city (Sunset district mostly) and they're figuring out now how they want to extend it.

I recall reading an article from a year or two ago stating that a lot of those cisterns are empty, and the overall backup water supply is woefully insufficient.

San Francisco has 66 fire trucks the last time I checked (I’m an SFFD volunteer). If there is a bad enough earthquake they will not come for you. The sheer volume of fires and service response calls will overwhelm them. It is your responsibility and the responsibility of volunteers in your community to make sure everyone is safe or triaged correctly the first 2 days (7 days better) after a disaster.

Neighborhoods will definitely burn but containment will be more of a factor of how windy it is and how easy it is to get to water.

> If there is a bad enough earthquake [the fire department] will not come for you. The sheer volume of fires and service response calls will overwhelm them.

Wish your comment was higher up.

If you're in the Bay Area: sign up for a NERT/CERT class, buy an ABC fire extinguisher, and know where and how to shut off your gas line. Assume the fire department will not get to you.

SF NERT: http://sf-fire.org/neighborhood-emergency-response-team-nert

If there are wide spread fires all around your building, I don't think small extinguishers from few uncoordinated volontaries would help.

In such event, I would do only one thing: run.

Fires start off small then grow. If you have a fire in your home and can put it out early by yourself, that's the best scenario.

I'm not advocating for anyone to run around playing fire fighter.

Many 1906 SF fires were intentionally set by the owners themselves, after the earthquake had damaged their buildings, because most insurance policies covered losses due to fire, but not those due to earthquake. So, if you burned your building down completely, leaving no evidence that it had been first damaged by the earthquake, you had a chance at being reimbursed. http://www.sfmuseum.org/1906.2/arson.html

It's not clear how much of the fire damage was due to intentionally set fires; there were also a significant number of fires started unintentionally when the fire department dynamited various buildings in failed attempts to establish fire breaks. (And, hey, you were still covered if your building was dynamited!) So, lets hope the SFFD is better in this regard than in 1906 as well.

The insurance industry claims that only 2% of the destruction was directly due to the earthquake; 98% was from the fires. https://www.iii.org/article/san-francisco-earthquake-1906-in...

> These buildings aren't unsafe because there's no choice, they're unsafe because the builders wanted to save money and the building code was too loose to stop them

Side note from a New Yorker, it always shocks me how careful San Francisco pretends to be with building codes and environmental and safety reviews and this and that, only to go on and systematically approve a series of terrible structures. (Not before showering them with ecological awards, of course [1].)

[1] https://en.wikipedia.org/wiki/Millennium_Tower_(San_Francisc...

I can't remember an article were an architect or engineer flat-out claimed that the Millennium Tower design was fundamentally flawed. I thought this article had such a quote but reading it again the claim is merely implied by the journalists.

The Millennium Tower was not flawed. It just unfortunately bumped into the law of large numbers. Even if all future towers in SF rest on bedrock, that still wouldn't reflect poorly the judgment of the Millennium Tower architects.

Frankly, New Yorkers should be more concerned about the seismic safety of New York buildings. In terms of seismic risk vs seismic structural resilience, New York may be facing a great chance of catastrophe than San Francisco.

New York also almost had a skyscraper blow over because no one bothered to calculate what the wind hitting it diagonally would do: https://en.wikipedia.org/wiki/Citigroup_Center

Regarding Japan: Nara has the Okumura Commemorative Museum [1] that has a bunch of little exhibits demonstrating how Japanese builders have been adapting to earthquakes. You can even look under the building at the foundation to see two of the methods in action - a crazy sort of "rubberized" slide pillar, as well as pillars just plonked on top of plates of teflon.

[1] https://www.okumuragumi.co.jp/en/commemorative/index.html

>How? By having some of the strictest building codes in the world, yet allowing development within those rules.

While this tends to be true for large, commercial architectural projects, it’s not true for residences. A big reason why almost no one buys used homes in Japan(and if they do, they tear them down and build new) is that building codes are very loosely enforced in this domain. If I remember correctly, structural calculations to resist earthquakes aren’t even required for buildings under 2.5 stories (i.e. just about all private residences).

It’s a commonly-cited misconception that zoning is what makes housing so cheap in Japan. Zoning is a factor, but in reality it’s because housing is built pretty fast and loose as a disposable commodity.

One of the reasons why it's harder to buy some used homes in Japan is because it's harder government backed housing loans for housing that do not meet earthquake standards. For older buildings it's also harder or not possible to get tax breaks.

I see things like this ("These buildings aren't unsafe because there's no choice, they're unsafe because the builders wanted to save money and the building code was too loose to stop them.") posted by people on HN quite often. As someone who has worked in the Bay Area on buildings, I read things like this and laugh. They come from a place of ignorance - anyone who thinks the building code is "too loose" has never successfully engineered something which complied with it.

I don't doubt you but could you elaborate? Thanks.

>> anyone who thinks the building code is "too loose" has never successfully engineered something which complied with it.

Perhaps, but have you compared it to the building code in Japan? I personally have qualifications in this area so I'm just asking.

Too late to edit. Why oh why did I omit the word "don't" in front of "have qualifications"? It was meant to be an honest question to fill a gap, not a weird confrontational thing.

Your comment would have more gravity if you pointed to specifics instead of touting some vague credentials and telling other people they're wrong.

> more gravity

Now that would be a serious problem.

Shallow dismissals make you look like you have nothing to say. The article posits one thing, then the OP gave a concrete example clarifying why one region is having problems the other doesn't. You then come in and just claim that the building codes aren't actually that loose while jerking yourself off about your experience in the matter, despite the article's position on unsafe buildings and the OP correctly pointing out that a region with similar problems doesn't have an issue with its high rise buildings that they allowed to be built.

Do you have anything else to say besides vague claims on your credentials and experience?

The fact of the matter is in a best case scenario, 10% of these buildings are going to collapse in a major earthquake. 10% of the newest, shiniest, up to code buildings. Best case, that's the engineering code, that's the spec. They are designed to collapse 10% of the time.

In reality many more will collapse, because some buildings are older and not up to code, as well as earthquakes generally just being weird and unpredictable.

Do you think 10% of buildings collapsing is a "loose" or "tight" code? 10% losses is acceptable in say, human habitation?

That's not how tolerances work.

You don't get exactly 10% failure rate, you have to demonstrate better than 10% failure rate which means going past that point. Further, failure is not 'collapse' failure is anyone in the building being unable to exit safely.

And again, they are aiming for a 9.0 that's a ridiculously large earthquake. An 7.8 similar to the last SF earthquake would still be a major earthquake, but vastly less dangerous by comparison.

A code can be both too strict and too loose but for different aspects. The article makes clear that the building code trusts skyscraper developers much more than makers of smaller structures. E.g., "A five-story building has the same strength requirements as a 50-story building."

The obvious counterproof to your argument is Millennium Tower, a 60-story skyscraper, which has sank at least 16 inches so far. Most of those one 1 side, so the building now tilts at least 15 inches at the top: https://www.bloomberg.com/news/articles/2017-02-01/who-will-...

They are still trying to figure out how to fix it. Doing so could cost well more than it took to built it in the first place. One proposed plan is to anchor it to bedrock on the low side, then let the tower sink further until it's about even before anchoring the other side: https://sf.curbed.com/2018/4/16/17242450/millennium-tower-si...

I honestly think it's reasonable that the building code gives skyscrapers a lot of leeway. Traditional building codes are very prescriptive, and that works better with (common and similar) houses than (rare and varied) skyscrapers. But the ongoing clusterfuck that is Millennium Tower makes it clear we can't just trust developers to get it right. As mabbo says, they have a strong financial incentive to cut corners.

Why would the strength requirement for a 5 story building be different from a 50 story building? Do the building materials behave differently when you increase a number?

The Millennium Tower is not a "counter proof" to my argument, although it has become a bellwether for exactly the kind of posters I call out in my original post. My argument is that people on here who do armchair structural engineer and make comments like "the building code is too lax" are ignorant and full of shit. All the Millenium Tower proves is that, even with a building code, it is still possible to get unintended consequences. To the rest of your post about skyscrapers getting a lot of leeway - they don't get any more leeway than any other structure, they just have the money it takes to invoke portions of the code that allow alternative analysis methods. It just doesn't make sense financially to do some of the complex modeling for a 3 story building to shave 5% of material costs.

If anything, the building code is more stringent on skyscrapers because it does require more advanced analysis. This analysis is NOT precluded for smaller buildings, it is just not worth the money.

> Why would the strength requirement for a 5 story building be different from a 50 story building?

Because if it falls over it won't just kill 5 stories worth of people, but 50 stories plus an unknown number of people in the building it falls on, as well as creating a giant civic problem?

> ignorant and full of shit

Oooohkay, buddy. At best you are playing semantic games here. If your theory is that the building code is just fine when things like the Millennium Tower can happen, then we disagree on what "fine" means.

> If your theory is that the building code is just fine when things like the Millennium Tower can happen, then we disagree on what "fine" means.

What portion of the building code do you think needs to be updated in response to the Millennium situation? In reality, the building code achieves a success rate far in excess of what engineers deem acceptable. It is unfortunate that the Millennium is a big glaring exception, but the fact remains.

Geotechnical engineering (the cause of the Millennium lean) remains a highly empirical field of engineering with uncertainties that dwarf structural ones. Hence there is a lot of built-in conservatism in design parameters like bearing pressures. But there are limits to the amount of information you can ascertain of what's below you that you cannot see. Looking back in retrospect and saying "well obviously they should have had longer piles" like many people have on the various threads about the tower is a pretty ignorant thing to say, and it shows that they do not have an appreciation of the engineering behind the decision. There a tens of thousands of examples of friction piles that have worked flawlessly. The engineering evidence appears to have supported the decision, but the in-situ soil conditions varied from their design model. This, in my opinion, is not a failure of codes, but an unfortunate, statistically rare outcome.

Engineers may deem this acceptable, but given that San Francisco taxpayers may end up footing the bill, I doubt the rest of us do.

I understand that the people who built it believed that they were doing just fine. But I also understand that there are enormous pressures to keep costs low on a project like this. The notion that for this building those enormous pressures had exactly zero contribution to this outcome is a claim I certainly wouldn't accept without evidence.

If this were really in the range of statistically expected outcomes, then a) there should have been a plan in place to fix it before the structure went up, and b) they would have had sufficient insurance coverage to pay for the fix. Neither is true. The ongoing gosh-golly-who-cooda-node routine plus the simultaneous festival of blame-shifting suggests to me that, at best, the business side of the operation went well beyond what an engineer would have done in the same place.

The correct fix to the building code here I'm happy to leave to the experts, just as long as they eliminate outcomes like this. If that ends up raising the costs for other fancy buildings stuffed with pied-a-terres for the wealthy even to prohibitive levels, I am entirely fine with that. Erring on the side of of earthquake safety while building near major faults in major metropolitan areas seems like a great choice to me.

> just as long as they eliminate outcomes like this

I guess the point I'm trying to make is that complete elimination isn't possible in a long tail statistical distribution. There will always be a point where we have to draw the line, and there will always be freak occurrences of undesirable outcomes. This is the crux of engineering - finding the correct balance between cost and safety. Perhaps the current state is not enough, but the wildly disproportionate amount of structures that are working just fine tells me we are close.

> If that ends up raising the costs for other fancy buildings stuffed with pied-a-terres for the wealthy even to prohibitive levels, I am entirely fine with that.

Building codes aren't only applied to high end buildings. Any changes affect the (badly needed) low end housing just as much as high end buildings. Just another thing to consider.

> Erring on the side of of earthquake safety while building near major faults in major metropolitan areas seems like a great choice to me.

No disagreement there ;)

The outcome I'm referring to isn't just the sinking building. It's the clusterfuck of a sinking building and years of chaos that could spill over into extensive costs for others. Long-tail outcomes perhaps can't be eliminated, but they can be 100% mitigated through planning and insurance.

Medical care has a much longer tail, for example, but many countries manage to provide full financial coverage and individual support in treating and managing even very rare diseases.

If you're saying this was a reasonably forseeable but unlikely outcome of cost-driven skipping out on the bedrock anchoring that neighboring buildings used, the I don't think it's too much to ask them to have an answer to the question of, "Ok, if your cheap solution doesn't work, what will you do next?" And if they don't have a good answer to that, I also don't think it's too much for them to not be allowed to build the thing.

Reply on news.ycombinator.com