Unrealistic Displacements in Frequency Studies?

[tluxon]

I'm doing frequency studies on a mass cantilevered/suspended by spring metal strips. I've been getting what appear to be reasonable frequency results, but grossly unrealistic URES Displacement values.

To do a sanity check, I've got a 4" long x 1" wide x 1/4" thick beam fixed to a vertical wall with a 2" cubic mass (approx. 2 pounds if steel material) merged off the free end of it. I've fixed the beam where the wall attachment would be and added a simple gravity load. When running a frequency study, the mode of interest is at about 59 Hz but is showing a resultant displacement of almost 5 inches.

I'm not familiar with this kind of result as my prior experience with frequency studies didn't reveal such unrealistic displacements. Can someone please explain this?

Thank you!
Tim

 

[corus]

The displacements are characterstic displacements and have no real meaning. You'd need to apply some excitation force or diplacement in order to calculate real values.

Tata

 

[eduardocamargo]

Just to complete what Corus've said.

No matter which software you're using in MODAL STUDIES (FREQUENCY), the results are:
- When the models reachs its natural ressonance;
- How it would behave in ressonance;

So we have:
- WHEN: is the frequency (in rad/sec or Hz) resulta
- HOW: is the deformed plot

All other values are merely qualitative.

Best regards,
Eduardo

 

[Scheffers]

As Eduardo says..

The HOW is very usefull, use the displacement-result and let it animate, you'll see in which directions you can make your design stiffer..

 

[tluxon]

So why is a displacement of 5 inches given to a beam that is only 4 inches long? Is it supposed to represent some kind of percentage or proportion of that particular direction in relationi to all the various reactions? If the 5 inches is meaningless, why isn't some unit of measurement other than inches used?

 

[KevinDeSmet]

True. True. Maybe it should be considered a unitless value? Would be less confusing.

Certified SolidWorks Professional

 

[AlgorDesign]

The displacement results are proportional across the mdoel, so that can be useful. But as previously mentioned, there is no true response (absolute displacement) being calculated in that study.

If you want actual displacement, you'll need to run a linear dynamic study type.

 

[kellnerp]

This is a beginners error. Modal displacements are always scaled to make the shape visible. To find the real displacement at a particular frequency you have to run a frequency analysis which will show the actual displacement and stress. A modal analysis yields eigenfrequencies, eigenvalues and eigenmodes. The eigenvalues are not displacements.

If you think about a modal analysis it is impossible to give real displacements because there is no load specified for the run.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

http://www.linkedin.com/in/engineeringtransport

"Node news is good news."