Windloads in high structures

Hi Everybody,

I have a question regarding windloads. First of all I was wondering how to quantify them. I found a formula which is quite generic however it allows you at least to calculate the Pressure for a given windspeed :

http://de.wikipedia.org/wiki/Winddruck

So far so good. Windspeed varies with surface roughness of the surrounding area (faster over e.g. water (low roughness) etc... slower in e.g. cities (higher roughness)) in general, which you can get from a windmap. As they also do vary with distance from the ground you needed to express this as a gradient in section. I was wondering how to set this up reasonably correct as deflections of structure differ quite a bit depending on how loads are quantified and applied.

I am dealing with a high (400 m) so far open (no facades) lattice structure, so what I tried first is taking a meshLoad Component which distributes loads on peripheral nodes of the lattice. This seems to work ok, however deflections for a horizontal loads is quite high approx 7m for 300 km/h windspeed (4.2 kn/m2) (not taking the gradient towards the ground into consideration), although it is a relatively dense tubular diagrid structure. I guess deflection should be not more than 50 cms ?

The other method I tried, as I was thinking that basically all the beams will take the windloads, to distribute loads onto the beams by multiplying diameter (cross section height) * Pressure per sqm. This is yielding similar results as the approach with the mesh. Guess with the beam approach would be easier to give the beams closer to the ground less load (gradient I mentioned before), I recon with the mesh load component you had to superimpose several loaded meshes or divide the mesh in load areas.

I think my question in the end is, what would be a reasonable deflection for a high structure like this in a design phase and also what strategies are there to improve deflection values (maybe apart from adjusting the shape and density of the structure). Would something like a mass damper already be needed in structure of that height ? If yes, could this be simulated within Karamba ? Are there certain height to structure weight ratios which should be achieved ? ( I am assuming steel as material at the moment) Also, are there Tubular Standard Profiles beyond 100cm diameter in the ProfilesFile ? I could not find any, that's why I am using my own definition for Profiles at the moment.

many thanks for reading,

best Philipp

Replies to This Discussion

Permalink Reply by Steve Lewis on March 20, 2013 at 12:19pm

For wind you need to consult an appropriate design guide such as ASCE 7-05.

Wind loading on a structure is a VERY complicated subject.

Currently there are two approaches.

A) A desktop or analytic approach of estimating wind loads using design codes such as ASCE7-05.

B) Hiring a wind expert to construct a physical model and and calculate wind pressures measured directly from a wind tunnel test.

A) will allow you to derive a site specific wind design pressure based on the height of the building, surface roughness, site location etc. Typically you then multiply this pressure by an appropriate co-efficent in the code for a given building shape.

The other atypical method it to use numerical approaches such as CFD. This approach is not yet accepted due to nuances such as of surface roughness.

Building deflection is again subjective. Doing a modal analysis and getting an idea of the frequency is better. You can increase the frequency by playing about with the building stiffness.

You need to modify the stiffness of the building to get the deflection down. i.e. play about with geometry, add extra members with stiffness, reduce weight, use material with higher elastic modulus etc

If you are getting a 7m deflection for a 400m building then I can right away that is way too soft. That is equivalent to 400m/7m or H/57. You want to be at least H/500 to H/1000 so aim for 80 to 40 mm.

Your wind load seems reasonable at 87 psf for that height of building

RSS