Normal modes analysies with 1D-Masses

Altair Forum User

Hello,

I want to analyze the normal modes of a complex system and want to simplify it by using 1D-Masses. I created a test model, which is attached at the end, to learn about the handling in Hypermesh. It is basically a plate with a block attached to it. I want to analyze the normal modes by replacing the block with a 1D-Mass, entering the moments, which I extracted from the CAD.

To verify my experiment I analyzed the normal modes of the whole assembly group. The results are my comparison data, which should also be achieved with my minimized model.

So after that I deleted the block, attached a 1D-Mass with rigids on the plate and entered the moments of inertia. The results differ a little bit from the original ones.

 

So my question: Is the way I minimized my model correct? Are the data correct and only differ because of rounding errors while entering the moments of inertia? If not, what is the correct way to do it?

Thank you in before

 

https://drive.google.com/file/d/0B1pCdqzVxCnXbWszZHU5VEpZaVU/view?usp=sharing

Rahul Rajan_21763

Hi,

We request you to have a look at Book written by Nitin Gokhale for Practical finite element analysis.

Chapter 14 & chapter 9 for Dynamic analysis and special element techniques.

Results correlation depends upon many parameter.

HyperWorks Student guides / ebooks - http://www.altairuniversity.com/free-study-guide-book/

Model file which you have attached contains one block with four ends fixed for modal analysis seems to be okay.

Regards

Rahul R

Altair Forum User

Hi,

 

Yes, the one block analysis is okay, but that wasn't the problem. The Question is, wether the model with the 1D mass attached to it is simulated correctly as the model with the block and the blok attached to it. I want to simplify the multi block model as a one block model with an 1D mass. Is this model correct? The results are not exactly the same, but it could also be because of rounding errors i guess? There are multiple models attached to my post.

 

I entered the mass coordinates and the moments of inertia, I am ONLY interested in the frequencies and NOT in the modal formes. I know, that the mass point cannot simulate the modal formes, but my question is can he simulate masses for eigenfrequencies?

 

The student guide does not really help...

Rahul Rajan_21763

Hi, 

 

Use CONM1 or CONM2 to apply mass.Please refer attached thumbnail.

For sure this will have effect on natural frequency.

 

Regards

Rahul R

<?xml version="1.0" encoding="UTF-8"?>

<?xml version="1.0" encoding="UTF-8"?>

Altair Forum User

I already used a conm mass in my model... Do you checked all my models?

Rahul Rajan_21763

Hi,

 

Since its a numerical method difficult to get the exact answer.By the way how much is the variation in frequency from FEA and experimental results?

 

Regards

Rahul

Altair Forum User

Hi,

as mentioned, simplifying a body with a mass element will have an effect on the eigenfrequencies. By using a mass element, you will only have 6 degrees of freedom / modes for this body. While the 'real' body has many more like bending, torsion, tension, 1st ... n-order. This means by putting a mass element to replace a body in modal analysis, you will neglect the elasticity of this body and consider it as 'rigid'.

 

A second effect might occur if you use RBE2 (rigid) spiders to attach the mass to your ground body: the RBE2 will stiffen your ground body and will additionally alter the natural frequencies. An attachment via RBE3 (soft) spiders would be more suitable in this case.

 

Jan