Wind drift - Can drift due to gravity loads be subtracted?

[EngrRC]

In ASCE7, wind drift is evaluated using load combination of 1.0 dead + 0.5 live + Service wind. If I run two nonlinear cases (A) 1.0 Dead + 0.5 Live + service wind, and (B) 1.0 Dead [w/o Facade loads], can I evaluate the drift as A-B?

My thinking is that wind drift is concerned with facade/non-structural partition damage, hence I can take out permanent lateral displacements due to gravity loads during construction.

In my case, most of the lateral displacement is coming from gravity loads due to eccentric floor plates / transfers / increasing floor area at upper floors.

 

[STpipe]

I wouldn't think so. The commentary on serviceability in ASCE 7 makes it clear that drift should be evaluated based on that load combination. To me it would be similar to the deflection requirement for beams. You have the typical live load deflection limit of L/360, however you can't simply ignore the effects of dead load and should also consider limiting dead+live load deflections to L/240.

You didn't mention what sort of structure you're designing, but if it's steel and you're running non-linear load cases to account for P-delta effects, do you have the stiffness reductions applied? Those aren't meant to be used for serviceability checks.

 

[hardbutmild]

It really depends on what you're trying to avoid. If it's the damage of non-structural elements that were added after dead loads, then you can look only at additional deflection.
I don't know how it's in ASCE, but in Eurocodes there is a check for this and also a limit for "visual user feeling". There is also a note that limits can be changed if an investor wants to - i don't know if this is the case in the US.

 

[phamENG]

Drift and vertical displacements aren't the same, so you can't subtract one from the other. Those gravity loads may induce second order effects in your frame when you consider drift from wind or seismic, causing the lateral deflection to increase.

 

[EngrRC]

Structure lateral system is RC walls with outriggers. There are planted columns supported on steel transfer trusses at mechanical levels.

phamENG, you may have misunderstood. I am referring to lateral displacement due to gravity loads (not vertical displacement) that's causing exceedance of drift. I find that if I take out the displacement component coming from construction stage, limits would be satisfied in the long term, but somehow authority does not approve of this.

 

[phamENG]

There should be no lateral displacements from construction. If there were, you'd have a crooked building.

 

[slickdeals]

@EngrRC - for buildings where gravity loading isn't symmetric, it is not unusual to end up with lateral movements resulting from it. When such types of buildings exist, you should consider running construction sequence analysis. ETABS (and other programs) have the ability to analyze the building as it's built one level at a time, unlike building the entire model and then switching on gravity. Once you figure out how much the building is leaning due to gravity loads from a staged construction analysis, you might be required to provide a compensation strategy to the contractor. As a simple example, if the columns on one side of the building are shortening due to offset in stiffness, you can compensate by building those columns up to a higher elevation so that as the columns compress they don't induce lateral sway.

You might be able to get some more feedback if you provide more clarity. But if gravity loads cause lateral sway, they must be considered. Especially in a tall building that is susceptible to creep, the gravity induced sway will continue to occur even after installation of glazing.

 

[EngrRC]

Exactly as you pointed out @slickdeals, the building is asymmetric and with alternating floor plates causing the lean.
The only effect that I am considering to take out is the instantaneous drift due to the structure's own weight (e.g, floor 'n' leans towards y direction, while floor 'n+1' leads towards x direction in alternating manner), as this will not be experienced by the facade (the primary reason why we're looking at drifts). Simply put, facade will be installed only after completing the structure so as the only movement will come from wind/earthquake and long-term effects.

However, in my experience, authorities do not agree with this approach.