Seismic Analysis Eccentricity (Mz) Problems

[Anonymous]

Hi everyone, I'm forced to open a new Thread due to my problems in seismic analysis with Robot.

I've made a simple structure to learn how to work with the seismic analysis and i've followed the webinar N°7 about seismic analysis with spectrum.

My main problem is with eccentricity, infact the results that Robot returns to me are different from what i expected and from what i've manually calculated.

In order:

1) I've set up my model and i defined a rigid diaphram trought rigid links

2) I've run a modal analysis without eccentricity

3) I've run a seismic analysis without eccentricity in X and Y direction

4) I've set main mode 1 for Y direction and main mode 2 for Y direction

In this model i've no structural (e0) eccentricity in x direction so i'll focus on seismic action in X direction. 

Reading the result i'm not agree with Mz that Robot returns to me, infact it should be: Mz = Fx * ey + Fy = Fx * ey (infact ex0 = 0)

As you can see in the pictures above my eo is equal to 0,56m , how can Mz be near 3 times Fx, this means Robot consider an eccentricity of Mz/Fx = 318.55/122 = 2,61 m that i can't find anywhere.

In Y direction Mz is correct and equal to 0, infact i have ex0 = 0

P.S. I've manually applied a point in the center of mass of each floor and i have applied to this point (as nodal force) the Fx that Robot returns to me and the result are different in displacements , stresses etc

[Anonymous]

In my opinion is similiar to what @mustafahesenow has exposed in this thread:

https://forums.autodesk.com/t5/robot-structural-analysis-forum/strange-stories-reduced-forces-result...

But unfortunately i cannot understand the solution. I've noticed that everything works fine with symmetrical structures without ex0 and ey0, in this type of buildings even the accidental eccentricity works fine. The problem is with non symmetrical buildings, Torsion values have completily no sense to me.

[Anonymous]

Please, there is no one that have a solution for this? I'm struggling to find the reason of these results.

I'm still doing example and i still have problem with unsymmetrical buildings.

[Anonymous]

I think the message of Mister CLAUDEO point out a real serious problem about the reduced moment of torsion obtained in “Results of stories”.

As a result of many tests, we notice that the calculated results are relatively far different from what we obtain by hand’s calculations and hence it comes up with greater efforts in columns...walls and foundations.

Please, could the technical team make a contribution here?

Thank you.

[mustafahesenow]

Hi @Anonymous

As you mentioned you need to apply the eccentricity from seismic force not from modal analysis..

I have searched about this and found out that this is the correct procedure required in all seismic codes 

See attached picture

[Anonymous]

Many thanks for your reply @mustafahesenow , i did what you said but the problem still remains. I studied this simple 3d structure and i still don't understand the results.

I started with a complete simmetrical biulding and i have no problems in results, everything is simmetrycal and G=R so Mz = 0

Then i rotate this column of 10 degrees due to have a non symmetrical buildings, and the results have completely no sense to me:

I expected to have Mz = Fy*ex + Fx*ey = 2,43*0,2 + 45,56*0,1 = 5,042 KNm.

I'm making a lot of examples and the results are always strange.

[mustafahesenow]

Hi @Anonymous

Send me the model Please.

[Anonymous]

Here it is @mustafahesenow, i share it here becouse i can't find a way to send files in private message. The result of seismic analysis have completely no sense, expecially the ones in X direction.

[Anonymous]

I'm still making example of seismic analysis to go deeply in the meaning of the results that Robot returns. I've run a sesimic analysis in X direction without accidental eccentricity and i've make an equivalent static load case with the same forces that Robot has given to me as results of the seismic analysis and applied in the Center of Mass of the stories.

1) Results of seismic analysis in X direction

2) Results of the equivalent static load case

As you can see in the pictures the applied forces are exactly the same but the results are quite different

[t.sautierr]

Hi cleo,

The article you mentionned tells about the excentricity for seismic analysis due to incertainities in the center of mass / center of rigidity.

For your building, it has a natural geometrical excentricity with or without what you set in the seismic analysis, I don't what you are struggling with. when looking

at the efforts in the bars, it looks good, your question is about the efforts related to floors?

[Anonymous]

Hi @t.sautierr and thanks for your interest. Yes my main problem is with the Mz showed in the results related to floors, it's much bigger compared to what i've calculated manually. 

[t.sautierr]

Cleo,

I have no time to explore deeply your model today but :

remember that : 

- comparison between modal analysis and nodal force applied is not equivalent for structure with more that 1 mode : in the seismic results, force displacement etc ... are the results of a combination of modal shapes (CQC or SRSS usually) so the force you applied manually is distributed amongst many modes in the modal analysis.

- floor results are simplified results, so the information can be "twisted", as you have highlighted in your example, rotating slighty one column as created excentricy : because it has shifted the center of rigidity regarding the main Axis, so it an ecentricity of rigidity and not of mass(or vica versa, depend the point of view). For floors, Robot simply calculates 1 center of mass, 2 center of rigidity, does 1 - 2 = e0, regarding the efforts it does M /V not more.

I you want to analyse precisely, you shall analyse effort in the bars to calculate reaction,etc ... floor table is simplified tool even it shall be correct.

[Anonymous]

Many thanks for your answer, i know the problem about CQC or SRSS analysis and of course i do not pretend to have an exact results about torsion, it just seems strange to have a such different results, i know the results of Robot are exacts becouse the base reactions give me an Mz close to what a calculated manually, I'm just curious about the meaning of the Mz showed. Thanks for your explanation about results for buildings, i've run analysis with just 1 mode in modal analysis and unfortunately "results for buildigns" doesn't show me any results in "Forces".

[t.sautierr]

You need at least two modes (one for X et one for Y), then it displays results for "buildings". With one mode only, it gives nothing in buildings but it gives for "floors".