0
$\begingroup$

I'm looking to do a frequency analysis of a fixed-wing drone. The goal is to calculate the first few natural frequencies of the fuselage to verify there is no resonance with the wing structural/flutter modes.

The battery makes up a large portion of the fuselage weight/structure and adds stiffness. What would be the best approach to calculate the fuselage natural frequencies that account for both the structure itself and the battery?

I'm using sim scale to do the FEA analysis and am wondering in particular what I should assign for the battery material to account for the correct stiffness. (note that I have the battery geometrically modeled as a single rigid body). I can make a custom material and get a representative density but not sure what I should use for young's modulus and Poisson's ratio.

Also wondering if there are any representative hand calcs that I can do to verify the FEA for this situation.

$\endgroup$

1 Answer 1

0
$\begingroup$

There exist at least two schools of thought on this, as follows.

CAD software is incredibly powerful these days, and there are programs that can take a CAD representation of an engineering object and perform vibration/modal analysis on it. You just buy the license, download the programs, learn how to run them and do the job. Then you compare the results of the model with the actual behavior of the object, to verify the model's accuracy and applicability to your specific problem.

But there were ways of doing this which existed before computational power got cheap, which I myself have done in designing sound-deadening enclosures for noise control when I was in grad school. You build an accurate prototype of the object, instrument it with accelerometers, and strike it with a hammer in strategic locations to excite the modes. The primary modes are easy to catch this way.

Note that in the CAD-driven solution method, you still have to build a model, instrument it, and run experiments with it to verify the model. The second, no-CAD solution, skips the computer modeling altogether and goes straight to the same set of experiments.

$\endgroup$
1
  • $\begingroup$ These days one needs a cad model to make the scaled physical model economically so they just do both, producing some numbers earlier. $\endgroup$
    – Abel
    Commented Sep 19, 2023 at 2:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.