what exactly is harmful about a little net uplift at a frame column? 4

[hippo11]

I always hear that when you are doing preliminary lateral design for a bldg, you should check that there's no net uplift at the frame column foundation, in other words, make sure that

0.9 D - 1.0 E > 0 and
0.9 D - 1.3 W > 0

But my question is, what's wrong if you have a bit more than zero? What's so evil about 2 kips of net uplift, heck, what's wrong with 30 kips of net uplift?

It's not like your frame column will fly up into space at 1 kip net uplift.

What's wrong with the foundation losing contact with the soil in an earthquake or hurricane, as long as your structure is still stable?

A frame column that has lifted up 0.05" off the soil still has about the same amount of tension in it as a frame column that is still in contact with the soil. It just seems like an arbitrary threshold...as long as you can show thru a P-delta analysis that you are still stable, you should be okay even with net uplift occuring...right?

 

[UcfSE]

Deformations are checked at service level with service load regardless of strength design philosophy. Even with your design hurricane, even with LRFD design for instance, you should still be checking deformations with service loads. If we allow inelastic deformations at service levels, we have violated our standard of care and the building code.

That being said, if you are in a 120mph zone and a hurricane with 3-s gusts of 120mph or less floats in, you should remain elastic.

 

[Atomic25]

If you're counting on load redistribution to the other frames, and the system works, then why even have that frame there in the first place? Almost sounds like you have a frame you dont need.

Can your diaphragm accommodate this redistribution (connections and shear capacity)? Are you going to have excessive drift? Are you going to rain brick down on people running out of the building? All things you must consider.

 

[civilperson]

Uplift at one column is fine, stability of the system must be maintained with a FS. apsix has the correct approach.

 

[haynewp]

After skimming through this thread, I will note that the unwritten standard at 2 offices I have worked at was to use the surrounding slab on grade for uplift resistance. This involved calculating the cracking moment of the surrounding slab and using its weight to help hold the footing down. But that is as far as it went.

I am not sure you have to remain COMPLETELY elastic under a design hurricane as long as you can prove the building can withstand the sustained load of the event. I guess you could go through the system and figure if the footing and slab lifted up off the ground, then there is force redistribution through the girders and frame above (if there is a moment frame above), subsequent hinging and redistribution into the columns etc., but it is really not worth it I don't think.

We can't really predict the actual uplift forces in a design event all that well and I think it is too much trouble to do all this, and my budget is too small as it is.

I guess my question would be what are the deflection/cracking/repair limits under a hurricane event? We have drift for seismic and serviceability deflections for all kinds of wind on elements and service wind drift, but what about design level wind? Without researching this myself yet, I think hippo may be technically right but practically wrong, if he is still able to prove force balance in the end for a design wind uplift event. (I have probably shaken the bees nest around here again).

 

[miecz]

The code requires the designer to prove that the structure is stable (1604.4). In order to prove stability, it seems as though one would need to begin by calculating the uplift deflection. Not knowing how to do that, I would not allow a net uplift under load combinations that include wind.

 

[SteveGregory]

Thought...
If you lived in a trailer/mobile home, would you tolerate any net uplift? Or would you like to have your home properly anchored down to resist any uplift/overturning?

 

[hippo11]

If I lived in a mobile home chances are I'd be too loopy on crystal meth to care about uplift.

Kidding!

 

[JAE]

Even if it was a double-wide?

 

[csd72]

hipo11 and 271828,

Your statements scare me!

What alternative load paths exist that will stop the building turning over? What will stop the building continuing to move (or even accelerate) once it has started moving?

We are engineers, and as such we need to justify our designs in a scientific manner, not just write them off saying: I can believe that...

I can only think that you are referring to the calculated net uplift rather than the actual net uplift (under the theoretical loads). If there are additional dead weights that have not been taken into account, then take them into account.

I have used 10 times the thickness of the slab each side to resist the uplift and have even added the downwards component of the internal pressure on this same area, but I have always been able to avoid net uplift.

csd

 

[hippo11]

CSD,

Redistribution of loads due to redundancy....if you look through the posts carefully you will see specifically what we are visualizing. I'm not saying that it would be easy to quantify, but certainly, no one is talking about believing it with no specifics, or writing things off.

Why do you use 10 times the thickness of the slab on each side? That seems a bit arbitrary...actually the portion of SOG weight you can consider towards uplift resistance should be the footing size in plan + a bit more due to the shear friction angle of the soil between the underside of the SOG and t/ftg.

 

[csd72]

hippo11,

Sorry if I was a bit blunt.

I see what you are talking about in the case of an internal column in a grid. It would only be applicable if you had a tiff element over that would transfer the stress elsewhere.

The 10x is based on the fact that slabs usually span 20 times depth for floors. The top mesh in a floor slab means that it has some ability to cantilever out beyond the soil pressure cone.

csd

 

[bkjd]

How many buildings have we seen that have failed due to global instablity from overturning, assuming no foundation issue such as liquefaction? Or for that fact how about sliding? Both in terms of earthquake forces. Uplift is a good mode of failure if you think about... why?

What happens when the frame column starts to lift off? Does this not increase the fundemental building period which in turns drops the "dynamic earthquake force". Remember these forces are cycling back and forth it is not as if you have a constant force pushing to the left/right causing the overturnng, as soon as the frame lifts off the force drops off the force changes direction and away you go the for the next few cycles.

food for consumption.

regards,

bkgd

 

[rt06]

During preliminary lateral design, your column in tension is a good index that your building is either too light or too slim that it might have stability problem. It might be fine but on the other hand it leaves you not much room to play in the CD design, and every body knows how much the building might changess. If you building becomes lighter or slimmer then you will run into big trouble to make it work. During the final design phase I do not mind tension exists in a few columns, as long as I make sure the stability of the entire building is fine, and I have enoguh dowels to transfer the uplift forces to the rock.

 

[wadakis]

Great thread...

In regards to seismic controlled frame design, the current codes actually makes uplift the yield mechanism. My frames are design to be elastic at 1.4E, however, uplift is only required to be checked at 1.0E. Does this not mean that my footing will uplift before any yielding (and energy disipation) occurs in my ductile frame beams?

Additionally, concern with uplift is different depending on the number of bays within the moment frame. In a single bay frame uplift of the end column results in complete loss of stiffness of that frame and therefore your diaphragm will most likely get nailed. In multi-bay frames, uplift of the end column results in loss of the end bay only and now the first interior column acts as the end column. This puts very little addtionaly demand on the diaphragm (a little drag force).

 

[csd72]

hippo11,

Now that I better understand your question I will give a much better answer.

It is theoretically okay to allow net uplift on a small number of footings under ultimate loads if you have analysed the frame to redistribute the loads to the other footings.

As this requires analysing the frame twice, and has a number of possible serviceability implications it is much easier to design the footings to avoid this.

I would also question if there is any benefit as the other footings will need the additional weight to take the additional uplift load.

csd

 

[hippo11]

interesting...at the bottom of pg 115 of the "2006 IBC Structural/Seismic Design Manual Code Application Examples"


...."Note that this indicates uplift will occur. ASCE/SEI7-05 does NOT require that foundation stability be maintained using strength-level seismic forces...."