Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

natural freq. for beam with non-uniform cross-section

Status
Not open for further replies.

jbw9

Mechanical
Sep 15, 2009
4
Hi,

I would like to model a cantilever beam in a program like MathCAD that I can modify quickly. It seems silly using FEA (Nastran) to model this type of system just to get the first few modes and mode shapes.

I have been able to do this for a simple uniform beam using the flexural matrix and mass matrix, obtaining the eigenvalues and eigenvectors. However, I am having a hard time figuring out the best way to tackle the non-uniform cantilever beam. The example I am working is a section of pipe, coupling, another section of pipe (diff. schedule), and a large mass at the free end. I need to model the details b/c the solution needs to be very accurate. PLEASE HELP!! Thanks in advance.

I have attached a schematic.
 
Replies continue below

Recommended for you

Why is it silly to use FEA for such a problem?
In theory, a distributed parameter model would be the most accurate. Unless you are familiar with solving coupled PDE's then FEA is the easiest.
Although, if it were me, the first thing I would start with is lumping the pipe sections and then solve the lumped parameter system.
From here you can get a "feel".
Then move on to the PDE modeling.

[peace]
Fe
 
also, depending on how that coupling is "stationed" and made of it potentially could male your results "way off" perse is not modeled.

[peace]
Fe
 
If you were programming from the ground up, it is numerically simpler to solve the problem by Holzer's method, or Mykelstahd/Prohl transfer matrix method. I have programmed the latter in excel myself.

But if you already have an FEA program...

=====================================
Eng-tips forums: The best place on the web for engineering discussions.
 
Agree, I'd use FEA to check my handcalc. Or vice versa.

Anyway, Rayleigh Ritz is another good approach for this sort of problem, it depends on whether you'd rather use iteration (Holzer) or differentiation(R-R).

Holzer is probably better suited to mathcad





Cheers

Greg Locock

SIG:please see FAQ731-376 for tips on how to make the best use of Eng-Tips.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor