Modal Analysis Results 1

[eegultekin]

I am sharing etabs model at attached file. As you can see from model, i am analyzing a structure consist of panels up to 8 m level and steel columns from 8 meter to 13 meter. When you look modal participating ratios you can see that at first modes all x and y direction is zero. And at x side, it joins at 38. mode. Can you check the model, what is wrong with it. Thanks for your attention.

 

[JoshPlumSE]

There have been some other threads on this forum about modal participation mass ratios (per the ETABS terminology). I'm not sure that anything covers your exact situation. But, I'd search for threads with "local modes", "lumped mass", "discretized mass" or similar terms.

https://www.eng-tips.com/viewthread.cfm?qid=517025

In a case like this, where the mass is basically associated with the walls, you're going to get a lot of "local modes" associated with those walls. When I say "local modes" I mean modes that don't involve the majority of the structure... but rather a smaller portion of it. Like the portion of a wall between transvers supports vibrating by itself without really activating the lateral force resisting system.

There are lots of ways to deal with this.
a) You might choose to try Ritz Vectors instead of Eigen Vectors.... Which it looks like you've already done.
b) You might try to manually create a different mass source with a "discretized mass" that will reduce your local modes. This is essentially where you remove the self weight from that mass source and instead manually lump that mass at specific locations that capture the main spatial distribution of the mass... But, also minimize the local modes.

If I get a chance this weekend, I might look for some training materials I had for a training class on dynamics I used to teach (6+ years ago). It was for a different program. But, the concepts of discretizing the mass will be similar.

 

[eegultekin]

Thanks so much. I am waiting your commments at weekend. As you say because of walls, reinforced walls join to the modes at last modes like 38-39th modes.

 

[JoshPlumSE]

Whew... I feared that I lost that old training manual when I got rid of my old computer a couple of months ago. But, I did finally find it.

Attached is a PDF of a RISA model (because I built the training manual back when I worked for RISA) where I talk about creating a load case that avoids the local mode issues of a weird floor slab and shear wall model.

Essentially, we're just lumping the mass at the tops of the columns and tops of the wall. So, where the vertical and lateral systems meet. That avoids the local modes.

FWIW, the discretization is very sloppy for this model. I would have preferred to do it the following way:
a) Create a load case that is just the seismic self weight of the structure applied as a vertical load exactly as it exists.
b) Temporary add rigid vertical supports at the tops of the wall and columns. Solve for the load case in part a).
c) Review the reactions from solution b).
d) Use the reactions from c) to create a load case which includes ONLY the reactions from c) applied downwards as a mass or load that will be converted into a mass by the program.
e) Remove the fake vertical supports you created in b).... And, run your eigen solution using the mass source defined in d). Voila! You should have a mass source that captures the spatial distribution of your mass, but which de-emphasizes the local modes that are not likely to be part of the main system to resist EQ loading.