ETABS User Wind Loads on Semi-Rigid Diaphragms

[HD-111]

Hi Eng Tips Community,

I am currently in the process of designing the lateral stability system of a 30 storey concrete apartment building. The main lateral stability element is a concrete core box with a few internal walls. Currently the diaphragms are all modelled as semi-rigid as there are outrigger systems and the floorplate is not a regular rectangular floor shape.

We have recently received wind tunnel testing results in the form of Fx, Fy and Mz forces which are to be applied at a specified point location within the slab/diaphragm. Upon inputting these loads into the model, we are seeing some unreasonably high tension stresses in a few of the internal walls which don't make sense to us. Upon further investigating, it seems like re-defining the diaphragms as rigid fixes the issue. It appears that ETABS does not distribute the loads correctly when inputting user defined wind loads at a single point within the floorplate when using semi-rigid diaphragms. My understanding is that with auto wind or earthquake loads, ETABS correctly distributes the lateral forces among multiple joints within the floorplate when using semi-rigid diaphragms.

My question is whether or not there is a way to properly input wind tunnel test point loads while using semi-rigid diaphragms while distributing the load realistically to the core box.

Thanks in advance!

 

[sticksandtriangles]

If you put in a single point load, yeah you will get some odd diaphragm results. If you make a rigid diaphragm, you get no results in your diaphragm.

If you have wind tunnel loading, you can make your input in ETABs as detailed as you would like. If you want down and dirty, apply the load to one side of the building. More exact, maybe play with breaking it into windward, leeward and sidewall pressures. See below.

You can modify the applied loading as well to be a trapezoid to achieve Mz forces.

Also, you could take the section cut tool for a spin, it was meant to help review diaphragm shear and moment results in etabs.

https://github.com/retug/SectionCutter

 

[HD-111]

Thanks sticksandtriangles, I'll give this a go and see how it turns out. For anyone interested or having the same issue, see CSI's response regarding user wind loads onto semi-rigid diaphragms below:

User loads are not recommended to be used with semi-rigid diaphragms since the program applies local rigid constraints to move the load to a nearby joint object and this could cause unrealistic out of plane stress concentration in walls, beams... It could be also more problematic if the user load falls outside of the slab edge or within an opening. Instead, we suggest that you assign rigid diaphragm or find a way to apply the user loads as multiple joint loads that are statically equivalent to the single concentrated load. Our developers are aware of this item to find a better way to handle user loads with semi-rigid diaphragm in future release of the program.

 

[sticksandtriangles]

To be a contrarian, I have not had any trouble with applying user defined loads to semi-rigid diaphragms. Hopefully CSI expanded a bit more than the quote that you posted.

User loads are not recommended to be used with semi-rigid diaphragms since the program applies local rigid constraints to move the load to a nearby joint object and this could cause unrealistic out of plane stress concentration in walls, beams

If you have your semi rigid diaphragm modelled with enough sub-divisions, I feel like this is a non-issue. I typically model a frame element with very low stiffness along the edge of the windward/leeward edge of the building and assign a frame load to mimic the calculated wind pressure load. This frame member is then meshed with the semi-rigid diaphragm to get a close approximation.

It could be also more problematic if the user load falls outside of the slab edge or within an opening

I'm not sure what they are trying to get at here. Of course, the user defined load should be modelled where the engineer expects load to be.

Our developers are aware of this item to find a better way to handle user loads with semi-rigid diaphragm in future release of the program.

What is the issue with applying user defined loads to semi rigid diaphragms? Did they explain any more regarding this? It seems to work for me.

Sample, frame with user defined load on semi rigid diaphragm:

Deformed shape:

Section cut results:

S&T -

http://www.re-tug.com

 

[Trenno]

I posted something similar a few years ago. Might be of use. Link

 

[Trenno]

It's wild that ETABS isn't set up for something so common (any tower with a set of a wind tunnel point loads).