Hacker News new | past | comments | ask | show | jobs | submit login
What really happened at the Millennium Tower? (practical.engineering)
336 points by chmaynard 68 days ago | hide | past | favorite | 210 comments

I liked the article but the last paragraph was just wrong. Implying that the public were being irrational for not buying apartments in the building, when the engineers said it was safe is incorrect. The engineers said it is currently safe but it was sinking so the safety would have to be periodically re-evaluated in the future. When you are buying an apartment, the value you get is not only its current value of living in it, but the ability to live in it for many years in the future and the ability to sell it or give it to your kids some time in the distant future.

Considering the crazy SF prices, especially for new condo developments, the future is especially important when buying an apartment in SF.

So it is perfectly rational for people to refuse to buy based on an engineering statement that the building is safe but may not be safe a couple of years from now.

Nevermind the fact that the sinking was already three times worse or more what the same engineers that are assuring it is safe said it would do. Doesn't instill a lot of confidence in your future predictions when your past ones have been so completely wrong.

I'm not buying the "we didn't know people would build and drill nearby" argument. It's downtown San Francisco for God's sake.

Then, there is

"By that time, the building had settled another inch and was out-of-plumb by more than a foot or 30 centimeters. That’s not something you could notice by eye and was probably only discernible to the most perceptive residents, but it’s well beyond the 6 inches allowed by the building code."

Twice the amount allowed by the building code sounds like a problem.

I find it strange that it would be quoted in a distance and not an angle. If at the time of writing, nobody anticipated a building this tall, the 6 inches might be very conservative.

Actually we use horizontal offset because it is more intuitive. Angles are deceptive for very tall buildings. You end up with a tiny angle eg 0.1deg that it sounds too small.

Also what people don't realize that the swing at the top floor of such a tall building due to wind may be actually about also another foot at worst case.

If the idea is to keep taller buildings straighter, then it makes sense. And that might be an important engineering consideration, considering that a taller building has more leverage.

It's usually proscribed in terms of inches per 100 feet of height.

For things like wind deflection, more than 2" per 100 feet is considered excessive.

Inches out of plumb are way easier to measure.

That's fine, since the height is well-known, you can convert that to an angle using nothing more than some high school trigonometry.

That would depend on how cautious the building code was in this case.

> Nevermind the fact that the sinking was already three times worse or more what the same engineers that are assuring it is safe said it would do.

It's not the same engineers. The first ones may well be wrong and the later ones right.

Maybe! Or they could both be wrong. If you'd like to bet that it's only one set of engineers, you can get in on the action for $0.7 - $3.4 million dollars: https://www.sanfranciscocondomania.com/condos/MILLENNIUM-TOW...

But given that most people buying a home are looking for a low-risk proposition, I get why prices dropped. Especially people buying condos; a big part of the point is that a lot of the hassles are taken care of for you.

> Especially people buying condos; a big part of the point is that a lot of the hassles are taken care of for you.

I sometimes wonder about that. For what it costs in ongoing condo fees, you could pay someone to provide exactly the same level of worry-free maintenance on a regular house.

The part that's worrying is the catastrophic cases. I had complex roof leak and while I had to yell and scream and threaten legal action, eventually the management company fixed it; my friend who had a similar problem in his house ended up wiping out his savings replacing the roof.

I can see that happening, but only when people don't save for such an eventuality. I was talking with a friend of mine a couple weeks ago and found out that his monthly condo fee is $1000. He spends $12K a year investing in the maintenance of the building. If a homeowner saves that same $1000/month into a maintenance account, they can put a very nice roof on their house every couple years. But roofs usually last 20+ years.

I certainly wouldn't pay $1k/month, lol. I'm sure there are bad deals out there, but the fundamental idea still seems sound, similar to insurance - pay a bit more on average in exchange for getting rid of the tail risk. If you rely on your own maintenance fund, you have to cover the whole of the worst case on your own; pooling that risk should make it cheaper.

Well, that probably wasn't necessary. That's basically what home equity loans were supposed to be for, back before people started treating their houses as ATMs.

He had the savings available (he'd been planning on retiring the next year), so I don't think getting a loan and keeping the savings would have made him any better off.

I believe the idea is, if you secure the loan against your house, and your house is beyond repair, you can default on the loan, surrender your house, and still have your savings.

Though I'm skeptical of the terms you'd get on a loan secured by a badly damaged home.

Well you're out the value of your house in that case, so I don't think that leaves you any better off. A loan might let you smooth out a sudden big expense, but it's not going to make it go away.

If a five-figure roof replacement wiped out his savings, how was he planning to pay for things in retirement?

I don't know all the details of his finances, but he'd worked in government all his life and had a pretty good pension. He also moved out of the city to a lower cost of living area, although I don't know whether that was in his original plan or not.

Ah, some government pensions are pretty generous, that makes sense.

Sell the house? I think that's a pretty common 'strategy', at this point.

And live..where?

Plus, if you sell the house with a bad roof, a smart buyer would insist on paying market value minus cost of roof replacement. If you replace roof, you’ll get market value. Either way, you’re going to lose the roof replacement cost.

I used mine as a way to deposit, and I'm pretty sure I'm not the only one. Banks have been pushing hard for people to mortgage against the property gain. It is easy to blame 'people' but banks were absolutely in on this and a major driver.

condo fees also cover walls out insurance, many utilities, neighborhood upkeep, etc. given you're already paying these yourself in SFH, it's not quite apples to apples to say full HOA fees would be ON TOP of all of that.

i'd subtract whatever your water/trash/sewage bills are from "typical" HOA rates in your area, and subtract typical difference in home vs HO6 insurance rates, then get to net to compare.

You could. At the same time, a new construction townhouse (and maybe even SFH) there is extremely limited maintenance work needed.

Plus, you don't need to be concerned with a HOA saying you can't rent your place out on whatever platform you choose.

Single family home and townhome HOA can, and often do restrict rentals. And short term rentals are restricted city wide

But how would anyone know which engineers would be guaranteed to be right? Who is to say any engineers would be for sure going to be right ahead of time, given how little we know?

Bur another (third?) set of engineers was also wrong, not anticipating the impact that their work to fix the situation would have on actually accelerating the problem.

It seems this is a chaotic environment that may defy the ability of experts to make accurate predictions.

  > Bur another (third?) set of engineers was also wrong, not anticipating the impact that their work to fix the situation would have on actually accelerating the problem.
The leaning tower of Pisa has a similar history: most attempts at straightening the tower only made it worse. I've heard this about another leaning building, also in Europe I believe, but I cannot remember where.

Thank you!

it's almost like humans don't know everything and are still learning.

and/or, details about the soil layer topology which weren't known when the building was built now are known.

> I'm not buying the "we didn't know people would build and drill nearby" argument. It's downtown San Francisco for God's sake.

Downtown San Francisco is NIMBY central, and it is quite surprising people were actually able to build something new there. It's a reasonable assumption that the zoning etc gridlock continues, and very little new construction happens nearby.

Actually, that neighborhood of SF is one of the very very few parts of the Bay Area that are not NIMBY central.

Grady Hillhouse is (per his /about page) a professional civil engineer. He just did a YouTube video on a screw-up that has launched a fleet of big-money lawsuits. He has a wife and little kid. If he is making sure that he won't be the target of any deep-pocketed legal action...that sounds pretty rational to me.

Also, I've noticed that we (computer professionals here on HN, roughly) don't always think that the general public is particularly rational in their choices about computers. Perhaps most professional civil engineers have similar feelings about the public being rational in their own area of expertise?

After the article about how moving the mouse around actually made Windows 95 load faster[0], and modern computers that do irritate users by changing settings on their own, I would say that engineers should probably listen to users more often.

0. https://www.extremetech.com/computing/294907-why-moving-the-...

> Also, I've noticed that we (computer professionals here on HN, roughly) don't always think that the general public is particularly rational in their choices about computers.

Hah, how’s that for an understatement. Most devs definitely see the general public much as Joe Wilkinson sees Fabio.

From my limited experience Civil Engineers tend to find the public frustratingly shortsighted - although that was water infastructure which yeah, is underinvested as evidenced by the lead stories, let alone the taste.

Yeah this topic is demonstrated in the UK. Long story short the UK failed to regulate building materials, resulting in lots of buildings being clad with inappropriate materials (really long story). A building burned to the ground (Grenfell) killing people. Now there are buildings all across the country with these unsafe materials, and they need to be replaced. There's a huge argument about who should pay - the government? It's their failure to regulate. The builders? They're ultimately responsible for what they built - but would go bankrupt immediately rather than pay? Individual owners? It's their property - but the costs are enormous, and they were sold something with the expectation it would be safe.

So whilst they try and determine who pays, these properties are now basically worthless, banks won't give mortgages, no one will buy them.

I'd say it's the builders' fault, but the lack of capital requirements are the governments fault. The gov should cover bankrupted build firms.

I very aggressively abridged that story, but the builders bought cladding material that suppliers told them conformed to the appropriate regulations. In reality the regulations weren't fit for purpose or enforced. There are questions around how aware of this they were.

Sounds like the banks are just dumb at the surface level. If they take a collateraled loan and pay you get a paid loan. If they take it and fail to pay you either get paid by the proceeds of the sale or real estate for peanuts compared to the cost basis to hold and sell for a way better rate later.

There's also the very real possibility of a massive special assessment on the condo owners to pay to fix the building that should be pushing down the property values.

Moreover, most people can't afford to buy a luxury condo without a mortgage, and banks won't give mortgages on condos where the HOA is in litigation with the builders.

I’ll buy one when one of those engineers does and goes to live in it.

I don't think they pay most construction engineers enough to purchase even one of the cheaper condos in that building.

Kind of like how high-end tailors can't afford one of their own suits.

Yeah, the Hammurabi Code…

The video doesn't mention that the tower is a lot of reinforced concrete, not steel as in most high rises. That makes it much heavier than other friction pile buildings which is at minimum contributing to the problem. IMHO I think it's insane anyone proposed not drilling to bedrock and that the city ever allowed it. In an earthquake zone bedrock piles should be an absolute requirement for tall buildings.

Last I heard the modified drilling procedure didn't help and the accelerated sinking hadn't fully stopped so right now there is no plan to resume.

When the pile drilling started lean was 17". In less than a year of that work it's now 22". By building settlement standards that's blindingly fast and makes me think the soil on that corner is in much worse shape than anyone assumed... possibly bad enough it won't hold in a large earthquake. It may be a more complicated interaction than just compression due to the load.

I'm not sure what they can do at this point. Have a bunch of crews get the piles installed as quickly as possible and hope you can do it before the sinking becomes dangerous? Given it is a concrete building rather than steel they probably can't do much to reduce the weight but moving everyone and everything possible out might help a little. To counteract the leak maybe they can drill on the other corner (that's always been part of the long-term fix) to get it to settle at a faster and thus more even rate?

Definitely a difficult problem.

I’m not an engineer but I am an architect and I have worked closely with engineers on skyscrapers. Drilling all the way down to the bedrock is risky. When you drill a deep pile you need to use bentonite slurry to stop the pile collapsing in on itself, the pile auger is withdrawn and then a long pipe is used to introduce concrete from the bottom of the the pile which should displace the slurry out of the top because it is less dense than the concrete. However, this process doesn’t always work and you get bentonite inclusions in the concrete which are found when the pile is tested ultrasonically afterwards. I don’t remember the stats but lets say it’s something like 1 in 10 piles fail the test when you drill to 50m deep bedrock through clay. You then have to drill two more either side of the original pile and bridge over the defective pile but these each have the same risk of inclusions, so there’s an increased risk the workaround will fail. As you can imagine this can get pretty complicated and expensive quickly. As an architect I was involved because we needed to have a backup plan for how the basement would be reorganised if there were a lot of piling failures requiring extra bridging structures.

Very interesting. Thanks for sharing. You make a good point that it’s not so simple as just drilling to bedrock.

It's not that complicated though.

There's nothing stopping the builders to get the bad column out and try it again. And again. Yes it's expensive as fuck, but this is still the foundation we're talking about. This (and the load bearing core) is where cost must come after safety.

Yes, naturally safety needs to be quantified and that just gets us back to cost again in a sense, but that's why there's a minimal safety codified (in the building code).

It's all about cost. There are many ways to get those nice columns in the ground, it's possible to drill them and put annular rings on top of each other (just like with regular wells) then fill it with concrete. More expensive than the bentonite slurry method? Sure, but it's still simple.

I have no idea why this comment would be downvoted other than disgruntled property owners.

The downvotes are likely because the parent clearly didn't read the entire article, and presented information in conflict with it. The engineering team is testing modified drilling techniques so they can resume repair work, this time (hopefully) without causing more accelerated settling.

This is all outlined in the article, and it's frustrating to see people commenting when they haven't read it.

I'm not sure either but I did find confirmation that San Francisco no longer allows friction piles for any high-rise buildings in areas with "infirm" soils. Piles must be driven to bedrock.

What a great idea to have a transcription. Much faster to read it. Then, we can decide if we might want to watch, which in this case, definitely yes. Fascinating story. Great summary of geotechnical engineering. There will be much more to the story to follow, it looks like.

Practical engineering videos are always worth a watch in my opinion, when I see them come up in my feed I almost certainly watch!

He also does mini demonstrations that show, for example, how weirs work, why sinkholes form the way they do, or how heat affects the structural soundness of concrete.

They are really simple, but help me better understand the world we live in.

Yes, a worthwhile YT subscription.

His Patreon page is https://www.patreon.com/PracticalEngineering/ – I send a couple bucks a month. There’s no special reward for supporting, but it’s nice to encourage people making excellent educational materials.

Brady from Practical Engineering writes the script/article for every video first and then reads it word for word as narration. It’s a nice consistent format and I appreciate that he shares the script/article in concert with the video.


Reading is so much better for skimming, searching for interesting bits, skipping to the conclusion. Having a transcript is fantastic.

Yes, almost always prefer reading to video. The exceptions are for things that are best explained with a physical demonstration, i.e. how-to type stuff.

Aside from the illustrations Brady shows during the video, he also commonly has physical demonstrations. And the videos are pretty short and content-filled.

I think the channel is well worth subscribing to and watching in full, the transcripts are useful to come back to afterwards, but it’s not 30 seconds of content being stretched out to 15mn so all the ads can fit in.

Those prices depend not only on buyers’ confidence in the building’s safety but also their willingness to be associated with a building that is regularly in the news.

He's leaving out a big factor that any condo owner understands: maintenance fees. I wouldn't take a unit in that building if it were free, because for the lifetime of the building there will be ongoing excess maintenance associated with every system because of the settling. Imagine the extra wear on the high speed elevators running out of true vertical. Imagine the differential stresses placed on pipes and boilers and tanks that expect to sit vertical. Imagine the effect of continually reseating the building's connections to local services like sewer and electricity. And imagine the effect of the HOA spending money on grand fixes while routine maintenance gets deferred.

In the near term, the effect of the tilt is negligible, but buildings like these continually suffer from imperfect, discontinuous maintenace under the best conditions; the Millenium Tower will forever be marked by every realtor and owner as maintenance-expensive (meaning vastly higher monthly fees) compared to all the other buildings around.

I'm wondering how plumbing & grey/black water could possibly withstand inches of shear or compression stress without bursting leaks, let alone feet.

My guess is they already have flexible connections with a lot of those on account of earthquake safety. You don't want all your pipes bursting and liquefying the ground during an earthquake.

Depending on where the connections are in the building, they may also not have suffered as much settlement as the most affected corner of the building.

I’m guessing since they expected the building to settle there were some allowances. So it is possible, but of course with a settling 3x of what’s expected I wonder too.

There is an accompanying YouTube video: https://www.youtube.com/watch?v=ph9O9yJoeZY

Grady makes great civil engineering videos.

You say accompanying video, and I say this post is a transcript of that video. Subtle, but still different.

Thank you, I felt like I was reading a transcript the whole time, I would have preferred to just watch the video

Any time you read "Hi, my name is ____ and this is _____" you know it's a transcript. Nobody (hopefully) writes like that for reading consumption. It's just so awkard when read.

There was another one of these here just the other day about processing cocoa from bean to chocolate. The written text kept saying things that implied a visual reference, but no images to support it. It wasn't until the end of the transcript before the video itself was embedded.

It's great that the transcript exists though. I never would have watched the video (video is my least favourite format for information of any kind) so I would have missed out. More YouTubers should do transcripts!

I will basically never watch a video unless it’s the only possible source of information for something I desperately need about how to repair something (which is becoming more often, unfortunately).

It amazes me that ANYONE would prefer to watch a video over reading, I literally cannot understand why that would be the case.

It would be great if prefer-video people could write something up for video-never people like me on why they prefer it! Then I could try to make a truly terrible video explaining my perspective. ;)

I also didn't see the video and just read (and enjoyed) the text. (Of course you are right, but I'm just saying it is kinda meaningless which came first or which is supposed to be the important one, different audiences :P)

You say transcript, and I say this post is the script of that video. Subtle, but different.

Have you ever seen a script? This is not formatted as a script in the least bit. Nobody that knows anything about scripts would ever call this a script. It is a transcript.

It's not formatted as a script, but it is a script. This text was written first, and then he made a video where he reads the text.

A transcript, while definitionally broad, is commonly used with the implication that someone has spoken in free-form (ie not reading from a script), and the transcript is the writing down of their speech.

>This text was written first, and then he made a video where he reads the text.

You know this how?

I direct messaged grady via youtube and that was his response

The image at the top of the link is actually an embedded video

totally missed the play button. it hides pretty well under that arrow.

Thanks for the video. I missed it and it's better than the transcript because it has schemas (the current solution is hard to understand without them).

I'm surprised the piles don't break due to the sinking on only one side? Can it not happen?

That's got to be one of the finest tutorial videos I've ever watched (and I watch a lot)!

He didn't even build any special contraptions for this video; his little acrylic demonstrators explain complex processes in a very easy to understand way.

I live in the tower, kinda spooky to read/listen to this. I can’t say all the news about the accelerated sinking has made me feel amazing, but eh, I got a good deal. Glad I’m just renting and don’t own a unit here.

There’s a pretty in-depth discussion thread on this forum with a lot of people who seem to know more than me and my SWE friends: https://www.eng-tips.com/viewthread.cfm?qid=470048

If you put a marble on the floor will it roll?

No, I tried this, and did not observe a roll.

We live on the corner of the building that is sinking the most, so I'm not sure why we don't see any tilt. Perhaps we need to be higher up to observe it?

Does it feel like living on the Titanic?

I live in a country that is eternally and increasingly sinking. Nobody is panicking. Same reason why nobody in SF is panicking about earthquakes I'm sure.

Is it common for buildings to list?

One thing I have wondered is - when the building sank 18 inches, did they have to add a step down at the entrance?

I lived in the building for a while and there is a bit of subtle ramp into the building but it's not quite a 1.5 foot drop because the direction (Try looking at Google Street View from before 2020 at 350 Mission St and 399 Mission St)

How was it? Did you like it there? It strikes me as soulless.

Yes this would be interesting to know. Maybe when the building sinks the sidewalk sinks as well?


IIRC, I had read somewhere, some time ago, that London's skyscrapers, built on a thick clay deposit, are at least partially supported by large and strongly-built basements which effectively act as rafts floating in the clay [1].

I have no idea whether this is feasible in San Francisco. One difference is that London is not prone to large earthquakes.

1. "London's skyscrapers float on rafts embedded in the clay." https://en.wikipedia.org/wiki/London_Clay#Engineering

That particular sentence on the wiki is uncited, so trust it with extreme caution. That said, yes, anything is feasible with enough money, and generally you pay an engineer to determine what method will cost the minimum amount of money.

That's effectively what the Millenium Tower did - their engineers determined that friction piles to the sandy layer would be the cheapest acceptable option. Then, due to factors they may or may not have been able to predict, it turned out to not be an adequate solution.

If they had known beforehand, the next cheapest option would have been chosen - likely end-bearing piles to the bedrock, like they are now retrofitting. Floating raft foundations are truly a last resort due to their expense.

You are right - based on the numbers given at the start of the article, and a density of clay around 1700 Kg/M3, if one kept within the footprint of the tower, it seems one would have to excavate ~100 feet down to create a sufficient volume (less, of course, if you can broaden the chamber beyond that footprint, but I guess the scope for that is limited - and, on the other hand, the landfill layer may be less dense than clay.)

Big skyscrapers in London use a similar friction pile technique. E.g The Cheesegrater sits on friction piles which I believe are ø2.5m x 25m long.

There’s a suburb of Vancouver called Richmond that has been built on an old bog. There are building limits and other codes to deal with because of it but I heard recently that the ground in Richmond is actually becoming more stable of all the piles being driven into the ground from new condo developments and other infrastructure.

This is generally because of heave effects (clay rising up after loads being removed for basement excavation), and results in significant raft slabs (as much as 3m thick) and tension piles which can be horrendously expensive.

When I was looking at SF real estate I was surprised to discover nearly half the housing market was built on landfill. And surprisingly few of those buildings are built on bedrock.

Loma Prieta earth quake (1989) was when I learned a bunch of the Bay Area landfill is also sunken ships that have been mudded over.

That’s especially surprising in an area known for earthquakes.

I was going to make a joke about San Francisco culture, but honestly this kind of thinking happens everywhere. Look at all the people buying or building in areas with flood potential that is high and increasing.

I believe there's a neighborhood in Houston that was build on land specifically left undeveloped in the past because it floods every time there's a hurricane.

Where there's enough money to be made, you'll find shady business people and corrupt politicians willing to risk people's lives and livelihoods for a buck.

Plus, the federal government will keep giving people enough money to rebuild on flood plains, so it makes financial sense to keep building in places that keep getting washed away.

Hello from the Netherlands!

At least there you folks had the sense to confront the problem and control it. In the US we've just been insuring people for floods so they keep rebuilding: https://www.usnews.com/news/national-news/articles/2018-07-3...

Gekoloniseerd from the sea.

Ooh, not just a word I didn't know, but a whole meme complex I didn't know: https://knowyourmeme.com/memes/gekoloniseerd

There's documentaries on the post 1906 earthquake rebuild where they were going to build to be earthquake safe but as time went on and demand increased that got thrown out the window.

Though at some point I think code enforcement returned, especially for large buildings.

The most at-risk buildings in SF are so-called "soft story" buildings, low-rise multifamily dwellings built on top of a parking garage or commercial space, especially those constructed before increased regulation (I believe in the 70s or 80s): https://en.m.wikipedia.org/wiki/Soft_story_building

I used to live in the former garage of a soft story building (they built a kitchen/bathroom/etc - it was around 1,000 square feet). It was a great deal because we got sole access to the backyard (it was a house). But after reading about how they relaxed building standards after '06 rather than increased them, I couldn't wait to get out - the house was built in 1918.

The landlady would always talk about how the house held up great during the '89 earthquake. But that didn't impress me considering the epicenter was in Santa Cruz.

It's pretty rare to have a house, or a building, build on bedrock in the Bay Area.

As a contractor with a foundation that settled it's nerve racking.

That said, very few foundations that settle need to be repaired. The average homeowner thinks the opposite though.

I stopped my settling by draining mine, and the neighbor's roof downspouts to the street.

Even without that fix, I doubt my house would have settled further.

It's true that most buildings/houses settle.

> The Millennium Tower’s foundation consists of a 10-foot or 3-meter-thick concrete slab supported by 950 concrete friction piles driven into the subsurface to a depth of about 80 feet or 24 meters.

When I see modern townhouses slab foundation instead of proper basement (which Archimed law favors as soil is just a very slow liquid in which building is floating like a ship until of course it's piles are driven into bedrock) I cringe at that cost cutting though it is ok for a 3 story building. I never imagined that it would be done to a skyscraper - no basement and no bedrock reaching piles. The story of Millennium tower reminds those buildings in Siberia on permafrost where builders cheated and did shallower than needed foundation - you can notice those buildings from a distance by twisted pagoda style roofs, etc.

> When I see modern townhouses slab foundation instead of proper basement

A mat or raft slab foundation - such as the kind used in the Millennium Tower - can be just fine for a small structure such as townhome. A mat/raft foundation is different than a regular slab on grade foundation in that it is thicker and has steel reinforcement running throughout, not just in the perimeter footers. [1] Its strength is due to the way the steel reinforcement distributes loads across the subsurface.

My 1600 square foot house is built with such a foundation and I got a close-up look at it during a recent major structural renovation and it is incredibly robust, even over-engineered.

Just because it was a bad choice for a skyscraper doesn't make it a bad choice for far lighter structures.

1. https://www.finehomebuilding.com/2019/12/30/mat-slabs-an-inv...

The lack of basement and associated parking makes the architecture incredibly alienating too. Almost everywhere you go, it's all parkades, garage doors, staircases and gates - absolutely insane.

Did you by chance read TFA? There are several references to a basement.

i couldn't find any description of the main tower basement - it all sounds like the main tower slab mat is basically near the ground level. The garage is said to be under the small tower.

There are other SF high-rises using slab, yet their piles reach the bedrock. Also Millenium Tower is reinforced concrete instead of typical for such high-rises steel - it was more "economical" - and due to the leaning/sinking stress the concrete there is cracking.

> the City of San Francisco and the design engineers were carefully monitoring the building’s movement as construction of the retrofit got started in May 2021. It didn’t take long to notice an issue. The vibrations and disturbance of drilling through the Old Bay Clay were making the settlement accelerate. The speed at which the building was tilting and sinking started to increase as the drilling continued. In August 2021, construction was halted to reassess the plan and find a solution to install the foundation retrofit safely. As of this writing, crews are testing some revised drilling procedures

It sounds like they still don't have a solution to fix the leaning tower of San Francisco.

Dumb question: why not drill out a couple inches of earth under the foundation on the side that isn't sagging to let it sag that way? Use a high pressure jet or something similar? I mean, those high pressure water jets they use to cut metal are cool.

The issue is getting it to settle evenly?

So you dig down on the sides and then use something: a long drill, a water jet, to open up small 1 or two inch channels. Are there high power lasers that can do this?

With the stresses, the building will crush out the small open space and... settle?

"The developer’s engineers and the City have shown that the building is perfectly safe through detailed modeling and investigation. And yet, the prices of those luxury condominiums plummeted with the frenzy of reporting about the settlement and tilting."

So the initial modelling was badly wrong, but new! improved! modelling show it's safe. The dumb public is being completely irrational. /sarc

buildings don’t get built in a vacuum

Well, not most. Yet. And sure, possibly ever.

(Go read the article / watch the video ... and a few others of Grady's. He really does do excellent work.)

You'd need a pretty big vacuum. My shop vac can hold about 20 gallons and fits in a corner of my basement, so I can't imagine how large one would need to be to fit a skyscraper in there. I guess that's why they call them Dyson Spheres though.

What about the international space station? (For some value of "built".)

It's an example.

It is also not most buildings.

(It is the most expensive building, however.)

My "ever" referred to whether or not most buildings might ever be constructed in a vacuum. Clearly, some are, and it's all but certain that more will be.

> It is the most expensive building, however

It might have been topped by "the Great Mosque" (~100 billion USD): https://en.wikipedia.org/wiki/Masjid_al-Haram

Interesting, thanks.

ISS is not in Vacuum. Its only roughly 400km up and experiences constant drag

I'm going to have to be a bit pedantic here. The ISS is at an altitude where the air pressure is 10^-9 Torr, or about one ten-millionth as dense than the atmosphere at the surface of the earth.

In laboratory terms, that's an ultra high vacuum which only specialized facilities on earth can achieve. I think that calling it Vacuum is perfectly reasonable, unless your standard for vacuum is deep interstellar space?

I originally wrote, before submitting, 'For some value of "built" and "vacuum"' but figured that was too pedantic :) I figured the definition of "building" would be the greater sticking point.

What is this "too pedantic" of which you speak?

Norman Foster championed the Boeing 747 as a great building in a BBC TV special.


I love this answer. I’m thankful for thoughtful people like you.

more of a ship than a building I would say, although its maneuverability leaves something to be desired

C'mon! It runs circles around virtually every ship ever built!

Prices for condos there range from $750K to $3.3M

I suspect they are undervalued, but I wouldn't live there. Not because of the tilting, which appears to be on its way to being fixed, but because it strikes me as a soulless sardine can.


>Fifty-two piles will be drilled along the north and west sides of the tower all the way down to bedrock.

If the bedrock is now supporting a significant share of the building, does it make it more or less able to cope with earthquakes?

I love the "unconventional alternative proposed by the developer's engineer". There has to be a story there. I'm imagining the scene in "The Martian" where Donald Glover has his eureka moment and has to convince a bunch of big wigs to try this totally crazy idea - "Let's drill basically to the center of the Earth to get some leverage on this thing"

After watching that story of the condo collapse in Florida, it gives me the creeps to think about going near that thing.

I wonder if they could run liquid nitrogen through new piles down into the weak sub soils to seize it up?

Freezing of subsoil is frequently performed during construction projects, for instance when building new underground tunnels in London. It sounds awfully expensive for a permanent installation, and raises questions about what happens if it's not maintained well or the power goes out.

Maybe they freeze the low side to allow the differential settlement to balance back out. I don’t think freezing is that challenging or expensive (on the scale of costs they are having to consider here) and would take weeks to thaw out. Also, the freezing need not be super deep or expansive, just at along original foundation pile to give them more “grab” in poor conditions.

Freezing could be used as part of a perm fix too. I bet there’s a way to sequentially freeze subsoils in such a way to use frost heave to actually push the building back into original position to augment the foundation. Or create a frozen layer in the mid-foundation and then pump in high pressure concrete grout below the frozen layer. Freezing water is a powerful thing…

Armchair civil engineering is fun

But it might be feasible to do it to solidify the soil as they drill through it to the bedrock.

Sure. Freezing is usually done to stabilize neighbouring structures.

Extending the piles down to bedrock level was quoted at $500m, so they ditched the idea for this project.

While lots of fun, that’s more of a pre-construction approach. Freezing is nice and slow!

Thinky me wants to empty it out and move in homeless, for as many years as it has left.

When I was homeless, I would have preferred that to living in my car. Even tho it was a roomy car.

Practical Engineering also has a great YouTube channel, too. If you liked this, check them out on YouTube.

It’s nice that the article provides the metric equivalent to the emperial values.

> The developer’s engineers and the City have shown that the building is perfectly safe through detailed modeling and investigation. And yet, the prices of those luxury condominiums plummeted with the frenzy of reporting about the settlement and tilting.

Well, yeah... Living in a tall sinking/tilting building would give me huge anxiety no matter what experts say

Settlement is one of the conditions that SF Bay realtors draw a very hard line around. I have on several occasions seen just the possibility of settlement kill deals and drive prices down by huge percentages. Similar to black mold.

The explanation that makes sense to me is that settlement is a problem that could be very costly and disruptive over a long period of time, and it's expensive and complicated to prove otherwise. So in the case of the Millennium Tower, they did in fact attempt to prove that the settlement was not going to cause major long-term problems, but that seemingly had no impact on the price behavior. I think with such an ingrained pattern, there isn't really a way to force buyers to think unconventionally; everyone "knows" that you don't buy in a building with settlement, and no one "knows" that there are circumstances where it's OK to buy anyway.

> building is perfectly safe through detailed modeling and investigation

but if they had this great model, which you would entrust your life to, why isn't it used to predict this settling in the first place?

That, and the fact that those "detailed models" have been wrong at every step of the way.

Conspicuous by its absence in the article is any mention of how much additional lean can be tolerated before occupancy of the building becomes hazardous.

I think it’s about 18 inches before it’s a seismic risk. They have a very small budget for more mistakes. (The precise number is in an SF Chronicle article somewhere.)

So code is 1/3 of that and the building is currently at 2/3 of that.

> Well, yeah... Living in a tall sinking/tilting building would give me huge anxiety no matter what experts say

especially, since the same experts said that it will barely sink at all.

Safe at standstill. How about seismic resilience? Surely a leaning building isnt exactly well tested seismically.

This is addressed in the article: “…the panel found that the building was completely safe, and the settlement had not compromised its ability to withstand strong earthquakes.”

So some people screwed up a long time ago on the initial design because they didn’t take into account the possible impact of neighboring projects. Ok, geotechnical engineering gets a mulligan on that one.

Later, thoroughly on notice that things were screwed up, they designed a solution to fix the problem. But they couldn’t predict that the vibrations from drilling would cause more sinking.

Now we’re supposed to believe they can accurately predict what the vibrations of an earthquake will do?

If this building is a pile of rubble, somewhere geotechnical engineers will be saying that it couldn’t possibly be predicted.

Here’s hoping the big one doesn’t hit while this exciting experiment in foundation installation is ongoing!

We’ll see

Someone confidently modeled the original construction too

And it sounds like they were pretty much right on if it is still safe, no?

Except presumably the initial modeling concluded that the building would not settle more than allowed by code, which we now know was not accurate.

remains to be seen if it is indeed safe. What other factors did they forget to include in the model? It's an unknown unknown, and doing an inquiry after the building's collapse is no comfort for those who lost value, or even lives if it were to happen.

You could say that about anything. The pros here are saying it is safe, and they said the same thing before.

Didn’t the engineers also find that it would only sink a few inches, but it ended up sinking over a foot in a fraction of the time?

"However, they cautioned that the movement hadn’t stopped, and further tilting may affect the building’s safety."

Doesn't it cause practical problems also, in addition to anxiety? Small problems, yes, but that accumulate over time. Water that accumulates in corners? Cooking with non-level pans is annoying.

It reminds me of that famous French comic about a (false) superhero called "Superdupont" in charge of protecting all things French. In one story the enemies of France steal the standard meter, and as a result all buildings become askew. I can't find it online right now but it's extremely funny.

In theory, theory and practice are the same. In practice, though...

Maybe if the engineers lived there...

> maybe 100 pounds per square foot of building footprint.

I guess I'm not going to see something like 4.8 KN/m^2 in my life time in daily writings in the US. //sigh...

SI is undoubtedly a superior system but in this particular example I can easily get a feel for what 100 lbs per sqft is like (a large dog standing on one floor tile, got it). 4.8 KN/m^2… not so much.

The rest of the world can't due to the reverse issue, notwithstanding also not knowing intuitiviely whether that's pounds-force or pounds-mass (I am guessing the former and the acceleration is factored in). These conventions that make it easier for you don't hold so well for an international audience.

Yes and no.

Before the current conventions (I am old enough to have studied in those times) pressure on soil would have been expressed in (still metric) in kgf/m2 or kgf/cm2 or - simplified - in kg/m2 that way (IMHO) it is easier to "visualize" pressure.

The 5 kpa would have been (roughly) 500 kg/m2, i.e. 5-6 people standing very near to each other within 1 square meter (to give you a quick term of comparison slabs in buildings are calculated to bear between 250 and 400 kg/m2, some particular ones up to 600 kg/m2).

If you imagine the soil as a liquid, with a density of around 2,000 kg/m3, double that of water, when you remove (dig) 1 meter of soil the surface below can surely carry (as it had done undisturbed for years) some 2,000 kg/m2 (usually much more), if you dig some 3 meters, you have surely 6,000 kg/m2 (that is the typical depth if digging for any "ordinary" building with a basement or underground garage).

The 11,000 pounds per square foot correspond to (roughly) 54,000 kg/m2.

This is a lot but seemingly not 100 times the "average" house, or - maybe better - the average house has a very high safety factor impolied.

Fair points all round, but if you specify something as 500 kg/m² I know that you mean 500 kg of mass per m² and so know to multiply by 9.81 to get a force and thus pressure. I know it can’t already be a force, because that’s a newton.

I’m not suggesting that a “mass per area” convention is bad, rather that pounds is so interchangeably used as a force and as a mass that it’s not immediately obvious to the layman (especially international and not familiar with imperial units) which is being used here, because both could reasonably fit in this particular case when using area as a denominator.

What I was trying to evidence was not metric vs. imperial, only about "visualizing" the whole stuff.

A Pascal by itself is not easy to visualize, a number of kg on a square meter (or for those used to it a number of pounds on a square foot) are IMHO easy to imagine, the 500 kg/m2 we were talking about can be visualized as 5-6 people standing one close to another or as a tank 1 m x 1 m filled with 50 cm of water.

As a side note, there is no conversion between kg and kgf, they are essentially the same thing, the 9.81 is already included, and it comes into play when you start using the (BTW proper SI) Newton:


To be fair, you can keep it SI and make it 480KG/M2 or 5 persons in a huddle :p

Throughout the entire article he also provides metric units for each measurement, thankfully.

The SI unit is Pascal. hPa is frequently used in weather reports in the US.

When being pedantic, at least write it correctly: 4.8 kN/m² or 4.8 kNm⁻²

I was pleasantly surprised that an American article had metric units with sensible conversions in most cases.

Engineers in America largely use metric day-to-day, for most circulations. Pressure/stress is one of the odd-ball situations where both Imperial & SI units are used (seemingly at random).

In what industry? In aerospace we used imperial for everything, including psi for pressure. NAS fasteners I think are all specified in imperial.

I've been following this pretty closely, both because I live in SF and an acquaintance owns a unit there.

The situation is very bad.

Around 2 weeks ago a city engineer for the first time mentioned "possibly dismantling the building", because logically you just can't have a skyscraper that's unsafe in the middle of downtown SF. It threatens numerous other tall buildings.

Based on videos I've seen of the Kobe earthquake, liquefaction would destroy the building even if tilting didn't decommission both the elevators and plumbing first. I wouldn't occupy anything in its shadow.

As a counterpoint, I know a couple who are both big-building architects for major firms who currently live worry-free in the Millennium Tower. I think the average person's perception isn't much use when evaluating things like this.

The comment refers to a “city engineer” so not an average person but presumably someone qualified to comment.

When talking about architects not afraid to live in it, I was referring to OP's last sentence:

I wouldn't occupy anything in its shadow.

I also did briefly look to see if I could find any reference to a city engineer proposing dismantling the building, but couldn't find anything there.

My understanding is it would be "safe", in that you're very unlikely to die, but would - with almost 100% certainty - be uninhabitable after a significant earthquake.

Ironically, that literally describes the intent of the building code, and seismic design philosophy of the past and present codes in the developed world

Would you mind sharing where you saw that mentioned?

An investment in the millennium towers is definitely safe and effective, sad to see there is some misinformation spreading suggesting otherwise

As someone who just heard about this engineering failure and checked wikipedia too to confirm, I wouldn't recommend my friends to invest in it. It could be profitable if they manage to fix the building, but right now I will bet on a demolition.

Those high tower are just nonsense, from a human and an ecological point of vue, especially for residential buildings. New buildings, should have no more than six floors

Sure. I happen to live in Hong Kong, where there is very little space. Shall we then just throw the excess population in the sea?

I'd be surprised if some rich people didn't buy units in this building just to insure them in case something happens and they can get a big payout.

How would that work? Buy a unit for $1.5M and sit on it, while paying insurance premiums. Building falls down and you get your $1.5M back after a huge amount of hassle?

Or just buy naked insurance premiums? Wouldn’t surprise me if that were a thing.

Insurers won’t insure you without an insurable interest.

I've got a plan for infinite money. I'll just buy a car, take insurance on it, crash it, collect insurance money. I can't lose!

Surely the insurance companies are also aware of this fact.

You'd be surprised what insurance companies overlook sometimes. An actuary told me one company he worked for accidentally sold an insurance policy for death due to cancer, but didn't ask applicants whether or not they currently had cancer. So that was quite expensive lesson on selection bias for that company, lol.

There was a freakanomics episode about someone taking out a bunch of life insurance policies against seniors in hospice.

Assuming you get everything you want and then some sure. But settlements are compromises - I doubt any of the homeowners at end of the day are in the black. But their losses are less than if they had not sued. Not a great investment by any means

Insurance companies win on average. So the premiums collected are probably going to exceed the anticipated payout.

And your plan only works if you can actually get insurance on a unit at this point.

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