Modal analysis with Ansys

[Saturnino]

Hello,

This is my first question in the forum, sorry for my english.

I am involved in some modal correlation between test and numerical results with ansys. As a first step I have simulated weldings as a common nodes between the welded plates, and I have concluded that generally the local modes obtained by calculation are higher in frequency than the tested ones. So I understand this simulation is too stiff, and a better simulation of welding introducing some rotational stiffness should be necesssary. I have been looking for papers or reference documents talking about this problem but I have not succed.

Please, could anybody help me with this issue?


Thank you very much in advance for your collaboration

 

[corus]

If your results show your model is too stiff then try using a finer mesh, ie. more elements, as too few elements causes the model to appear overly stiff.

ex-corus (semi-detached)

 

[Saturnino]

Thanks for your answer.

I have tried a finer mesh and the results were equivalent.

 

[flash3780]

You may also take a close look at the type of elements you're using in your mesh. Lower-order tetrahedral elements tend to be overly stiff.

 

[Saturnino]

Thanks for your answer.

You are right, I knew that.

In fact I have used shell181 elements, and also I have tested hexas elements ( not tetrahedal) but the results remain similar.
Regards,

 

[Twoballcane]

I'm not an ANSYS person, but how are you constraining the plates? Fixed would be stiff and simply supported would be less stiff.

Tobalcane
"If you avoid failure, you also avoid success."

 

[TERIO]

How far off are your frequencies? Do you have any sketches or pictures of what you tested and/or of your model?

 

[Saturnino]

Thanks for your answers.

As a first step I joined the plates with the cantlivers on their welding positions by common node, so coupling the six dof. After that I tried coupling just the translations and not the rotations, and the results on the local modes of the plates were better ( lower in frequency). So I think this the way to follow but I would like to know a little bit more why this is happenning and if somebody has experienced the same or knows about papers or bibliography talking about that.

Thanks.

 

[Twoballcane]

Ahhhh, yes, that is a common mistake. If you fix the ends, you will get small defelctions which will give you higher frequencies. With simply supported ends, you will get bigger deflections, which will give you lower frequencies. You can see this if you do the analysis and calcualte the frequencies using the two different constraints.

Tobalcane
"If you avoid failure, you also avoid success."

 

[Saturnino]

Do you know bibliography or paper talking about this problem?

 

[johnhors]

Being basic theory, then any Mechanics of Solids text book will cover this subject.

http://www.Roshaz.com

 

[Saturnino]

I suppose you mean that is a basic theory that a boundary pinned condition is less stiff than a clamped condition, this is not the problem.

My problem is depending on the type of welding you have the stiffness of the junction is different and this cause a different dynamic response of the structure ( basically on the local modes). What I would like to know if there is any book or paper talking about how to manage that in a similar problem.

 

[Twoballcane]

Sorry Saturnino, there are none that I know that talks about this in particular. However, this would be a great topic for my next white paper for the next engineering symposium. If the committee accepts it and I get to present, I may send it off to Sound and Vibration publications to see if they would put it in their magazine.



Tobalcane
"If you avoid failure, you also avoid success."

 

[Twoballcane]

Oh, and also, you may want to repost your question in eng-tips vibration forum. You may get better results.

Tobalcane
"If you avoid failure, you also avoid success."

 

[hacksaw]

are you modeling the weldment? or just joining the plates?

the weldment is not a continuation of the plate, with some change in the elastic properties, and depending on the finishing, extra metal.

 

[hacksaw]

in the interest of stimulating you search, the following link

https://bib.irb.hr/datoteka/83867.cim_999.doc

addresses some of the details of a weldment analysis

good luch