How can we be sure foundations don’t move over time? I understand that if is resting on solid rock, it will pretty much stay in place, but what if is resting on something less solid?
Is there a point where the foundation is so large that it practically does not move at all, or do all foundations move to a degree and we just have to live with it?
There is a great deal of experiential data over time, and research into the subject. Foundations do move if not secured properly. Take a look at the Manhattan skyline; the tallest buildings are located where bedrock is fairly easily accessible. But there are techniques to build large structures on looser terrain that involve multiple deep piers or pilings.
https://civiltoday.com/geotechnical-engineering/foundation-engineering/161-types-of-foundation
Experience. We know to watch out for things like frost heaving: depending on where you live, building codes will require a footer that goes far enough below grade to assure the ground below it won’t freeze and heave. We also know to watch out for expansive soils, which can change their volume/shape when their moisture content changes. Other soil conditions can be problematic, too. Some years ago I oversaw a construction project at my workplace that involved significant excavation down to about 20’ below grade. The more the contractor dug, the more peat and marl they found, neither of which make for reliable foundation support. In the end they removed a lot of extra native soil and had to backfill with a type of soil better-suited to a sturdy foundation.
Sometimes the foundations move, but we’re ready for it (sort of):
The island had been predicted to sink 5.7 m (18 ft 8 in) by the most optimistic estimate as the weight of the material used for construction compressed the seabed silts. However, by 1999, the island had sunk 8.2 m (26 ft 11 in) – much more than predicted. The project became the most expensive civil works project in modern history after twenty years of planning, three years of construction and US$15bn of investment. Much of what was learned went into the successful artificial islands in silt deposits for New Kitakyushu Airport, Kobe Airport, and Chūbu Centrair International Airport. The lessons of Kansai Airport were also applied in the construction of Hong Kong International Airport.[21]
In 1991, the terminal construction commenced. To compensate for the sinking of the island, adjustable columns were designed to support the terminal building. These are extended by inserting thick metal plates at their bases.
And sometimes, despite our experience and best efforts, we drop the ball completely:
And as an example of what can happen when the foundation isn’t properly designed or built, look at Millennium Tower in San Francisco, a high-rise residential building that was completed in 2009 and shortly thereafter started to tilt and sink. The fix is to involve sinking new, deeper pilings below the building. (Though I have no idea how you drill pilings below an existing building.)
So you backfill some of the loose soil with a more solid soil. But there is still loose soil around the solid soil (I assume).
Is the idea that eventually it rests on a really big area of loose soil and therefore won’t move around?
Soil mechanics is moderately well understood. There’s no such thing as a rigid foundation. The idea is to control the bearing pressure and the stiffness of the soil such that the movement of the structure is kept to acceptable limits. Acceptable limits depend on the intended use of the structure.
Bearing pressure can be reduced by using larger foundation systems (footings, raft foundations, piles/drilled shafts), reducing the load (excavation of a basement reduces the net load on the subsurface since you’re removing the weight of the overburden), or increasing the strength/stiffness of the soil by excavating to a competent soil, compacting the soil, mixing chemicals into the soil, or grouting.
The main difficulty in this process is adequately knowing the soil properties underneath the structure. What generally happens is that engineers will drill a hole and take samples of the soil to test in a lab or perform some in-situ tests to determine strength/stiffness. Soil has a lot of variability though and testing is expensive so you end up with only a vey small % of the relevant soil that is actually tested. Engineers then extrapolate and then measure deformations during and after construction to confirm their design assumptions. They’re not always correct and then some remedial actions need to be taken.
My bon fides: I no longer work as a civil engineer, but I was (and am still for a bit) licensed as a professional engineer and have a MS in soil mechanics.
The whole of Mexico City is sinking and the foundations shift by inches per year. As I understand it, many years ago, they pumped out all the groundwater. The soil and clay that are left are compacting quite rapidly and irreversibly. I have read that there are large civic buildings with huge staircases leading down to buildings that used to be at ground level.
Here’s a great video on the Millenium Tower, what the original foundations were like, why they failed to keep the tower stable, and what is being done to stabilize it.
I was gonna post the same video; Grady totally fucking rocks.
Another approach, if the soil is unsuitable, is called surcharging. Essentially you pile a huge amount of dirt on top of the existing dirt, and over time it displaces the water and generally compresses it. After that, you move away the extra dirt, and what remains is relatively more stable.
Grady has a video/article on that as well:
In Chicago, bedrock is almost 100 feet down, so we spent our first century experimenting with different kinds of foundations that wouldn’t sink too much. A particular problem was uneven settling, as when the design called for a tall bell tower on one corner. We tried a bunch of piles (for lots of friction) and floating foundations and grillage foundations that sort of combined the two. Eventually we settled on drilling down to hardpan (at -40 feet) or bedrock (at -100) and pouring concrete pillars called caissons. Near where I live in the South Loop, however, wandering of the Chicago River many centuries ago seems to have left the subsurface hardpan more difficult to rely on during drilling, or something, because many of the new buildings here use piledrivers to push steel I-beams deep into the ground. That’s noisy and annoying for adjacent buildings, who fear vibration damage, but fascinating to watch. They chalk the length on the piers and somewhere between 90 and 105 feet, the ringing of the steel with each strike will suddenly turn dull. They give it a couple more hits for good measure, then move on to the next one. A welding crew follows and cuts the I-beam off near the ground and caps it with a connection to the rebar mat in the ground floor slab.
Venice was built over the centuries from about 400AD in the middle of a swamp/lagoon. the foundation for the buildings are pine logs pounded deep into the mud.
If you are visiting Venice, you will note that the floor of St. Marks, built almost 800 years ago, is very wavy. Many of the buildings along the canals you can see have big metal plates with bolts on them - which I believe are for long metal rods through the building to stop the walls from crumbling outward as the foundations shift.
Sh*t happens, people make do.
Here’s an update on the SF Millennium Tower:
https://apnews.com/article/san-francisco-cba26c1927148e980aa2cfe0b068a25e\
At the current rate without a fix, the Millennium Tower in just a few years could reach a 40-inch (1-meter) tilt, which would be the point at which the elevators and plumbing may no longer operate, said Ron Hamburger, the engineer.
“The building does continue to settle at a rate of about one-half inch per year and to tilt at a rate of about three inches per year,” he told supervisors last week. “It is doing this whether we are conducting work at the site or not.”
Some article I just read said the problem with that tower was that attempts to fix the issue (by drilling down and pouring additional pilings) had aggravated the situation with the drilling allowing the ground to shift. This increased the tilt instead of stopping it. So, for now they’ve paused further remedial work. (“Back to the drawing board?”)