Applying seismic excitation

[Yara Adam]

Hello Everyone,
I am using ABAQUS to Apply a seismic excitation on a cylindrical structure (silo),
the thing is that,
I run two dynamic analysis for this cylindrical structure,
1- the first analysis is with modal dynamic step.
2- the second analysis is with Implicit dynamic step.
in both analysis I applied the same acceleration in the Boundary condition (as an earthquake excitation).
while the first analysis exhibits very reasonable behaviour (vibration and displacement shape as shown in vedio 1) , the second one exhibits very strange behaviour under the accelerogram (as it just continue moving in one direction. So the behaviour in general is wired, as if we push the structure in one direction as shown in vedio 2).
so what could be the problem,
Should I expect the same results from the two analysis under elastic conditions,
[URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1656342602/tips/vedio1_lqkzhf.mp4[/url][URL unfurl="true"]https://res.cloudinary.com/engineering-com/video/upload/v1656342624/tips/vedio2_ypkyrp.mp4[/url]

 

[FEA way]

Make sure that everything in these two analyses, apart from the step type, is the same or equivalent. Especially when it comes to the boundary conditions and loading. Maybe you confused the DOFs or used different amplitude, for instance.

 

[Yara Adam]

FEA way,
Thank you so much for your response,
actually the two models are identical apart form the step type.
I have applied the Boundary conditions as shown in the photos.
in short, for the model with the step of implicit dynamic, I disactivate the Boundary conditions for the step of implicit dynamic, and applied another boundary condition which is totally fixed except for the horizontal DOF along which the horizontal acceleration is applied.

 

[Yara Adam]

the same problem is discussed here but no solution

https://www.eng-tips.com/viewthread.cfm?qid=344248

 

[FEA way]

What are the two preceding steps - what kind of analysis procedure do they use ? Is geometric nonlinearity enabled in the implicit dynamic step ?

Can you share the .cae file (or files) with these two models for further testing ?

 

[Yara Adam]

the preceding steps are,
1-initial,
2-General static (to apply the gravity )
3-Frequency (linear perturbation for modal anlysis).
4-Dynamic Implicit (to apply the acceleration)

both models are totally linear, NLgeom is off always.
can you write you email address to share the cae.

 

[FEA way]

Check the Abaqus documentation examples "Analysis of a cantilever subject to earthquake motion" and "Double cantilever subjected to multiple base motions". Both include a comparison of transient modal dynamic and dynamic implicit procedure for a case involving base motion. You can download the input files from there and see if you haven't forgotten about some important difference that needs to be accounted for when changing from one procedure to the other. I would also remove the first general static step for now until you find out what's wrong.

Unfortunately, we can't share our e-mail addresses here because it could attract spammers.

 

[Yara Adam]

Dear FEA way,
thank you so much for the valuable information,
this is my Email address, (samisytel@gmail.com)
Kindly send me a message so that I can have your contact,
the system did not accept to upload the cae file here, However, I can send it by Email.