Shear Distribution % to Walls / Frames 2

[ilyas415]

Hi all!

I am undertaking an analysis on a large building with shear cores and moment frames.

I need to understand how much %Shear the walls are carrying compared to the frames, is this possible?

There is a way to show pier forces in ETABS (Display Tables➜Results➜Wall Results➜Pier Forces).
However, this method gives only in-plane and out-of-plane shear. This quickly becomes very troublesome since I need to know how much shear the walls are carrying in terms of the X- and Y-direction.

Any advice?

Thanks

-------------
Ilyas

 

[Trenno]

Hi Iylas,

What codes are you working to? Some codes have commentary about minimum percentages and dual lateral stability systems.

Is this a seismic zone? Or are you only concerned with wind load distribution?

In any case, I think you try the following steps:

- Set Rigid Diaphragms to your floor slabs as this will distribute the lateral load based upon stiffness of the lateral stability elements.
- Tick the box to make ETABS calculate your Center of Rotation (COR).
- Run the model and check for instabilities.
- Extract the data for the COR of each storey using (ctrl+T).
- Set up a new 'unit' load pattern for each direction (X & Y) with 'User Loads'.
- Apply a single 100 kN point load at each storey's COR, for each unit load case.
- This is to try and minimse any torsional effects distorting/amplifying the shears in your walls.
- Sum them up, so you know how much base shear you're expecting (ie # of storeys x 100 kN).
- Run the model and check for instabilities.
- Check your Storey Response Plots to make sure the shears you're expecting are being captured properly in the model.
- Assign each length of wall in each direction an individual pier label.
- For each direction export the pier forces (crtl+T) for the unit load case.
- Sum up their lowest storey shears to get the component of total base shear taken by the walls.
- The rest should be taken by the moment frames.
- Voila, you should then be able to work out your percentages from there.

Hopefully this gets you going in the right direction.

 

[ilyas415]

Hi Trenno, sorry for the late response. I didn't check the site over the weekend.

I am using Eurocode for this project. And the site is in a seismic zone.
Wind loads aren't a major concern.

Thank you for the great suggested steps! I will try this hopefully today or tomorrow!
Will let you know how it goes.


-------------
Ilyas

 

[Agent666]

Name each pier to differentiate it in each direction, i.e. include an X or Y or something similar in the names. Export pier forces to excel and you can sort it appropriately using filters or via formulas.

I'd disagree with Trennos post in one item, adding 100kN to each storey is not going to be correct for a seismic load distribution. You should be taking some level of overall load and distributing it appropriately over the height (for example based on equivalent static distribution). If using a modal analysis you can scale the base shear to give some workable base shear like 1000kN, which will let you determine how the load is distributing.

Also if you are using user loads applied at the centre of mass you will need rigid diaphragms, if you have semi-rigid diaphragms then you will need to use user coefficients and define these for each level appropriate to the acceleration at that floor.

 

[ilyas415]

Thanks Agent666.

My piers are not orthogonal to the X- and Y directions because the building shape is curved in plan, kinda like the example below:

I understand that the exported pier forces will be the in-plane for each individual wall.
So that means i need to resolve those back into X- and Y-components right? It seems a little bit troublesome because in our case we have a very big development many shear cores. But if it has to be done then it has to be done. I'm mostly doing it as an internal exercise to gauge how much shear force they are resisting, but anyway I will see what happens.

Thanks!

-------------
Ilyas

 

[Trenno]

Agent, yes I agree the 100 kN unit load approach may not necessarily reflect the true seismic load distribution, but it's a starting point to get a feel for basic linear load distributions.

Alternatively, just use the overall base shear coming from the ETABS auto-generated ESA seismic loads for this exercise. The 100kN unit load approach just took away that additional complexity and took away any torsional effects that may come about from eccentricities between CoM and CoR.

Ilyas, I assume your moment frame is simply a perimeter frame connecting the edge columns? Do you need a perimeter moment frame?

Here's a very alternative approach... how about you replace all of the shear walls with pinned columns (say a column in each corner junction) so you can get the gravity distribution working nicely. Therefore the pinned columns won't contribute to the lateral stiffness of the building. Find a few beams/columns on the lower levels and record their major axis moments for a lateral only load. Now, add the shear walls back in and check the beam/column moments again. Use this difference to then find out how much less the moment frame needs to work. That would give you a feel for how the shear walls contribute to the building's lateral stability system.

In any case, I'd be expecting something like a 75:25 split between the cores and the moment frame. You may find the moment frame contributes much less than 25%, but I'd design it for a minimum 25% share to be on the safe side. Keen to hear from someone like KootK who have more experience with North American codes and dual lateral stability systems. Maybe I'm totally off...

 

[ilyas415]

Thanks guys!

It has been a bit of a messy week for me, but I think I will give the stuff a go this Friday/weekend.



-------------
Ilyas