Comment by gerdesj | Hacker News Reader

Comment by gerdesj | Hacker News Reader

Parent Full thread

"The building was constructed in 1981 ... building was sinking"

Sinking isn't always a problem - look at Venice! I studied Civ Eng at Plymouth Poly (now a university, Devon not MA) back in 1989ish. Back then concrete cancer was all the rage ... and gabions (but I digress.)

The failure of the building in Miami was catastrophic and not gradual. I don't think that monitoring would have worked but it is possible that stresses/strains built up over time - we'll have to wait and see. If it was gradual then there is a good argument in favour of fitting buildings like that with strain gauges as many bridges are. I do a lot of IoT stuff and I estimate that something like £20,000 would get you strain gauges and a monitoring system for a structure like that. It might be £10,000 or £100,000 but a building that size in Miami is worth quite a lot. I'll bet rent on a flat was something like $2,000pm or more.

Inspection of foundations is a tricky one. I am not a Civil Engineer but I trained for it but I don't have X ray eyes and nor do Miami's engineers. That's absolutely the wrong way to go about this sort of thing. I'm not even sure what an "inspection" even means.

There was a fundamental error made at the initial design stage in my opinion. I don't know what but probably relating to the parameters of the ground that was built on. By that I mean that the architects/struct engrs. were promised certain properties of the geotechnics/geophysics - that basically means they were told how the ground would work in holding their structure up.

When you design a structure, it needs to support itself and crucially, the thing it sits on needs to be able to support that structure. You also have to work to a timespan, so you use standard tables for wind strength, earthquake etc. You work to a 1 in n years events.

This building collapsed catastrophically in seconds without warning. I think this will become a real lesson for Civ Eng and not a simple "I told you so" thing and certainly not "shit concrete".

The Millennium Tower in SF has been sinking[1] since it was erected. Tenants have sued and now there is a fix proposed.

It all happened because the builder didn't want to go 200' to hit bedrock and instead cheapened out.

[1]https://www.sfgate.com/news/editorspicks/article/SF-s-sinkin...

I watched them build a 22-story high rise on landfill (mud/sand) in the SOMA district of San Francisco—it was right next to my building. After demolition of the old structure (a 3-story apartment) and excavation, they did the following:

- Pump water out of the excavated ground. It was a slow drain, maybe 50 gallons per minute, but it ran for months.

- Drive dozens of 90 foot steel pylons into the site with an enormous pile driver.

- Pour the foundation over the pylons.

The Millenium Tower is more than double the height of that, at 46 stories, and it’s right by the water. But they used the same length of pylon. Mind-boggling.

Can you build a small replica of this with a bucket of mud, sand, etc…?

This is good point. If you are willing to go deep enough, can you build anywhere?

Technically, probably yes, but the math could get ridiculous. Essentially you want to go into solid ground, and if you're in the ocean, solid ground might be miles below sea level. And reclaimed land doesn't change that. Also, if the material higher up shifts, your supports need to be able to survive the ground shifting around it, which means you need not just depth, but girth.

It would be very unusual for a city's headlands to be literally miles above solid ground, especially for the carribean, which is extremely shallow (hence the hurricanes).

Take a look at the Florida keys, where you can find bedrock after a couple of meters. You can even go out a mile into the sea and hit bedrock after 6-8 meters:

https://pubs.usgs.gov/pp/2007/1751/professional-paper/figure...

I wouldn’t know whether it’s very unusual, but https://blogs.egu.eu/divisions/ts/2019/06/17/in-search-of-th...:

“Digging down in Amsterdam, it takes about 1000 meters until we find the first solid rock”.

There are more or less solid sand layers at much lower depths, though. That’s what buildings stand on.

It would surprise me if other cities in the west of the Netherlands were different.

Keep in mind that bedrock in this context is limestone and oolite.

This rock is not an ideal foundation material. http://www.mgs.md.gov/geology/geohazards/engineering_problem...

Harder rock is deep, around 1,300m. (4,000ft). https://ufdc.ufl.edu/UF00000245/00001/20x

> especially for the carribean, which is extremely shallow (hence the hurricanes)

The Caribbean is nowhere near shallow, take a look at a marine chart or just the seafloor topography on google maps. The Caribbean is thousands (often over 10K) of feet deep in most areas.

Also, this has nothing to do with hurricanes, which originate off the west coast of Africa.

Also, the Florida Keys are not in the Caribbean.

There are many points south of SF against the Bay where the solid ground is down far enough that they instead drive piles until they find enough resistance.

Oracle's HQ is still standing. But the (East) Palo Alto 4 Seasons Hotel, whoops.

Although it was reported that it had experienced more than expected settling, didn't the 4 Seasons open? Are there still problems?

I vaguely remember that they went back and drove more piles, somehow, until the building was stable that the exterior glass facade no longer shattered on a regular basis. However they did the work, it wasn't visible from the outside.

> Sinking isn't always a problem - look at Venice!

This seems like simultaneously survivorship bias (there have been plenty of buildings in Venice that have collapsed or have been irreparably damaged) and a good example of the dangers of sinking (how many billions on repairs and massive engineering projects to save the city?)

A civil engineer told me "settlement isn't a problem, differential settlement is a problem."

Yes and no.

Both statements are correct, but differential settlement is not an "instantaneous" problem, when a building, particularly a reinforced concrete one, is subject to differential settlement it deforms itself following the movements of the foundations or of the soil under the foundations, and this normally allows for several centimeters of deformation (in practice a building as seen from a side is made out of squares, when the foundations on one side lower, squares become a rhomb/diamond). All the rigid elements in the plane of the square (please read as walls, be them brick or other material) are forced "out of their natural shape" and start developing cracks.

Nornally settlement is a process that is very, very slow, years, months or weeks, during which the cracks begin developing, long before the structure collapses.

This must have been a form of failure of the soil that happened in a very short time, more similar to a landslide than to the kind of settlement common in soils (usually sands) near a waterfront.

I never implied it was a fast problem, nor did I make two statements.

The engineer made two statements:

1) settlement isn't a problem,

2) differential settlement is a problem.

Both are correct, but in this case neither settlement nor differential settlement appears to be involved in the collapsing of the building.

Most likely the brittle failure was due to corrosion of the steel reinforcing. The corrosion due to chloride ion intrusion. Chloride ion intrusion because it's on the ocean.

Or socially, because fixing the problem was an economic non-starter for the condo association. Or a political non-starter depending on how you squint.

I'm learning that condo associations are essentially monetizing the structure as it was built how ever many years ago, and have just about no interest in long-term maintenance or value.

I'm no engineer either, but it seems logical to me that a building would not even notice mild to moderate uniform sinking, but that differential sinking could split the building in half.

Also, I guess that differential sinking can be detected before catastrophic failure if the sink rate were measured on two or more points on the building. If the south side is sinking at 2mm/year while the north side is sinking at 0.2mm/year, that sounds like a recipe for disaster that you don't need xray eyes to see.

Reply on news.ycombinator.com