Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

SF Tower settlement Part III 18

Status
Not open for further replies.

dik

Structural
Apr 13, 2001
25,685
thread815-412357
thread815-470048

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik
 
Replies continue below

Recommended for you

But the whole point is to not overload the piles. I think. There's supposed to be a maximum load of 1300 kips.

If you have to raise the pressure to account for the increased downward force from the building, that downward force still HAS to be passed through to the piles. You can't lessen that.

A problem seems likely when/if the downward force of the building overpowers the support ability of the 18 piles.



If I jack up my car using a hydraulic jack (as my father DID do, a few times), I lift up a corner of the car, off of the ground. If the downward force from the car is 1000 pounds, then the base of the jack pushes downwards on the ground 1000 pounds. If I pump the jack, I raise the internal pressure in the jack. But the car still pushes down at 1000 pounds. And the jack still pushes down 1000 pounds.

Pumping the jack handle may well move the car, but it does not change the downward force of either the car of the base of the jack. Which must be equal.


spsalso

 
...but if the load applied to the pile by hydraulics is greater than the actual load from the building, it will have a tendency to raise the building at that point... for better or for worse, I don't know, but I suspect if something doesn't fail, it will be for the better. The problem with moving something that has moved over a period of time is that there can be unintended consequences... often find that happening with historic brickwork. If you put more force on the jack than the car does... the car accelerates upwards. In this case, the building will accelerate upwards, albeit slowly.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
I think one point of confusion about the jacks is that dik is saying on this kind of jack you can set a hydraulic pressure that should be maintained, which translates to a force that the jack applies. This differs from a commonplace car jack, where you pump the fluid in to raise the car, then close the valve, whereupon the jack becomes a solid object (more or less)... unless an excessive load blows a seal in the jack.

So in the Millenium Tower jacks, these become devices to apply a calibrated force to the building, while allowing the building to react by rising, staying stationary, or continuing to sink, albeit at perhaps reduced rate.

Do I have that right?
 
That has been my understanding. The jack pressure can be varied to maintain the max 1300kip load.

Another thing being forgotten in this conversation is the built in fuse system. Loads over 1300 Kip would fail the fuse to limit the load on the piles and mat. See Pyke’s paper posted by Walnuts Jan 7.
 
What I missed is that there are (staying with the easier for me to visualize car jack) TWO jacks involved. There's the old original one that has a bit of a leak in a seal (thanks gwideman). So it's slowly sinking. "We" figure if we take some load off the jack, the leak might stop. So we add another jack. It's only a small jack, and can't take the full load. However, we pump it up to slowly take just SOME of the pressure/weight pushing down on the other jack. And because we have a pressure gauge and an underpaid apprentice to watch it, that allows us to put in just the right amount of lift to take some load off of the other jack, while not blowing out.

The TOTAL downwards force of the car (building) remains the same (until you fill it with helium). But it's now distributed differently.

This fuse thing. Looking forward to learning more about that. Unlike a circuit breaker, a fuse is a one-time thing. So it seems that if the fuse "blows", the whole Fix is then disconnected and totally useless. But expensive.


spsalso
 
almost, except the small jack, has to be as large as the big jack... thanks for the explanation gwide... I hope it works, but don't know. The US does not need to rival Pisa with a Leaning Tower, albeit a tad bigger. Actually the Pisa Tower was leaning as it was constructed... maybe this one was too.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
Spalso: you said: "What would be nice is to add the pressure readings on the jacks to the Tilt-O-Meter." For some weeks they have been shown on page 24 of the Tilt-O-Meter. Separately, in a presentation developed by the SGH staffer who actually does all the work, developed I assume for the EDRT, it is explained that after increasing the load in the jacks in five increments to approximately 550 kips, the jacks were "locked off", likely just by closing a valve in the hydraulic system as someone has mentioned above, because the same presentation does not appear to show any mechanical lock off mechanism. Key photos from this presentation are included in Josh Porter's YouTube video. Interestingly, the loads measured by the load cell then dropped off over a couple of weeks to something less than 500 kips. At the request of Jaxon van Derbeken, Hamburger was asked about this by the DBI, and he basically said that this was all as expected, which was of course blowing smoke. A smart structural engineer friend of mine who has long experience with large diameter steel pile as used in offshore applications, has suggested that this fall off in the measured load likely results from some load redistribution down the pile below the bottom of the casings at about 100 feet. With the installation and load application sequences that they used, the piles likely had some resistance in skin friction in the 100 feet of so interval of Old Bay Clay when the loads were first applied, but this would "relax" over time (because the platy clay minerals relax) and more of the load is carried in the Lower Alameda formation, and maybe the "bedrock", although I doubt that much of the load actually gets to the bedrock. This would cause minor shortening of the piles since a greater load would be carried through to 100 feet of Old Bay Clay. Since, as noted above, the jacks are now more or less rigid, this slightly reduces the load. By itself this should not have a big effect on the Tilt-O-Meter because it appears that excavation of the trench along Fremont Street (now completed I think) and the water table going up and down, and, most importantly, some bending upwards of the edge of the mat, have more effect. Overall, it is a very complicated system and no matter how many Ph.D.'s you have doing calculations, it is all a bit unpredictable. We will have to wait a couple of months for the water table to stabilize to see what the more permanent effects are, but it seems more than likely that any recovery of tilt is going to be only a small fraction of the additional tilt that they have caused by careless construction practices.
 
Ah, the "fuses". More research pretty much says the threaded rods are the "fuses". They're supposed to break if the pressure down towards the piles gets too high.

What's the plan for when/if the "fuses" break? It surely looks like the "fix" is over, since the "fuses" connect the piles to the foundation. No fuses--no connection.

Just wondering.


spsalso
 
Spsalso: the threaded rods are supposed to be very stretchy, and they may not actually break even in a major earthquake. It is possible that the rods would pull out first, as suggested by Josh Porter. But the bigger problem is that no-one knows. As far as I know the design team has a fancy structural model, but they have never applied vertical excitation to it. I think the Hamburger just refers to the current minimum code requirement of applying a nominal static vertical load. It does make you wonder.
 
There is a simple reason the current design is probably O.K. The six piles that are completed on Mission Street have been tested to a certain load resulted in the reduction of Millennium Tower's tilt. See Figure No: 099-06A of the Millennium Tower weekly report.
 
The fact that it is settling so much is not a good sign. In addition, I don't know what effect the eccentric support will have during a major seismic event. In particular, the supporting soils have a high risk of liquifaction during a major seismic event. I'm not so confident myself... but I'm not really experienced in seismic events.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
Hmmm. It is true that the tilt to the north has been very slightly reduced, and the tilt to the west even less, but this appears to be largely because the north-west corner of the mat has been bent up a bit. But the original point of the Fix was to obtain a significant recovery of the tilt and any recovery to date is just a tiny fraction of the additional tilt that has been caused by inappropriate construction techniques. I don't think this is OK!
 
Has any of the post-construction effort indicated an end-all-be-all tilt that absolutely cannot be hit? Obviously the current status is not what you want out of a building, but at what point does this transform from a major screw-up to an extreme safety hazard that could warrant teardown? I know that's not likely to happen, just want to gauge the scale of the tilt-o-meter graphs to see what the target final tilt is. (Just reading through the thread now, I haven't been following the story too closely).
 
The Tilt-O-Meter is currently "flat lining", so your question may be moot.

But it IS a good question, one which I am curious if the San Francisco Building Department has ever considered. And what their conclusions might be.

In a reflection on the ka-wality of design and construction for this building, note that Millennium Tower is one of the buildings that had windows blown out in the windstorm of last month.


spsalso
 

That's great, now to address any seismic issues.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
Not just the seismic issues. The tilt is now significantly worse as a result of inappropriate construction procedures and so far the "rebound" that the design team had predicted does not seem to be happening. So, you still can't practice your putting in your multi-million-dollar condo! And, while no-one knows for sure, there are some reasons to expect that long-term settlement and tilting will resume.
 
Has it been considered that rather than stopping the settlement, they may be controlling the settlement?
If they new system provides enough support in addition to the original foundation piles that may be enough to stop settlement.
If the down hill side is pressured to the design maximum, but the up hill is held at quite a bit less pressure, they may be able to let settlement on the up hill side to right the building.
Once the building is vertical, the hydraulic pressures may be equalized.
If the building continues to settle, albeit much more slowly, the hydraulic pressures may be balanced to keep the building vertical.
If the intent is to slow and control further settlement, the system may work well.
But, what do I know.
Hoping for comments if my simplistic thoughts are too simplistic.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!
 
waross, your thoughts are not too simplistic, but the history of this fiasco is complex! Originally the design team said that after they propped up the north and west sides of the Tower, the south and east sides would continue settling and in something like 8 years the building would be level. Then they switched to saying that when some of the load was transferred to the perimeter piles, the south and east sides would "rebound" in the short term before overall long-term settlement resumed. But the only thing that can be said with certainty is that no-one knows!
 
Did they ever determine if the original foundation was OK to start with?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik
 
I think that determination has been well and truly made, by the performance of the building.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor