modal analysis 1

[CR250]

I'd like to better understand the difference between two types of vibration problems, wave propagation problems and structural dynamics problems, and what is being measured in a typical experimental modal analysis. <br>
<br>
Quoting from CONCEPTS AND APPLICATIONS OF FINITE ELEMENT ANALYSIS by Cook,[In wave propagation problems the loading is often an impact or an explosive blast. The excitation, and hence the structural response, are rich in high frequencies. In such problems we are usually interested in the effects of stress waves. Thus the time duration of analysis is usually short and is typically of a wave transversal time across a structure. A problem that is not a wave propagation problem, but for which inertia is important, is called a structural dynamics problem. In this category, the frequency of excitation is usually of the same order as the structure's lowest natural frequencies of vibration. ]<br>
<br>
It is common in experimental modal analysis to use a force hammer to excite a continuous structure, to quantify the modal shapes, frequencies, etc..., which is information associated with a structural dynamics type of vibration problem, where inertia is important. Isn't this type of excitation going to produce stress waves, for which accelerometers will pick up the associated stress wave displacements, as well as those associated with gross movement of the structure in a structural dynamics fashion ?. <br>
<br>
It would seem that how one excites a structure with a force hammer would have a significant impact on whether the structure vibrates in a wave fashion or in a structural dynamics fashion. For instance, striking a cantilevered beam on it's end would seem to likely excite nothing more than stress waves ?. In running a typical modal analysis, doesn't one need to have an idea of the structural dynamics vibration shapes or displacement directions first in order to excite the structure in that fashion ?.<br>
<br>
Also, doesn't an FEM analytical modal analysis only yield the structural dynamics solution ?. If that's true, then wouldn't we expect to see differences between FEM modal solutions and experimental modal solutions with force hammer excitation, aside from the differences due to normal engineering assumptions and experimental error?. <br>
<br>

 

[viblab]

The whole area of &quot;modal analysis&quot;, analytical and experimental, involves standing waves. Musical instruments are a good example, especially strings, which are easy to picture. I.E. the string is always distorted in the same shape only varying plus and minus and how much. An ocean wave is a feely example of a traveling wave. Using a hammer to excite the structure is a &quot;quick and dirty&quot; way to get the job done. The hammers used usually have rubber or plastic tips to keep the excitation in the low frequency (0-100 Hz.for rubber, 0-500 Hz for plastic). More precise modal experiments employ an exciter system imparting measured random forces. Even with a hammer a force gage is embedded between the tip and the body. The complex ratio of response acceleration-to-force as a function of frequency, called frequency response function is the data base from which mode shapes, frequencies, and dampings are extracted. Much of the early and ongoing work in &quot;modern&quot; experimental analysis was, and continues tobe done at U of Cinncinnatti by Dave Brown and Randy Allemang. I'm sure they could send you all kinds of stuff about the details. I do this testing at Boeing in Huntington Beach on Launch vehicles and sattelites. Our customers do not consider a finite element model enough basis to fly until it is test verified (and corrected if necessary). It's really neat stuff. We also do pyrotechnic testing, which involves mostly frequencies to 100k Hz. and up and mostly traveling waves, wherein proximity to the source is a major factor because they don't travel very far. Hope this helps clear things up for you.

 

[5ubtle]

It is my understanding (not sure) that in experimental modal analysis you hit the structure with a hammer and then wait until the transients die out. The software looks for frequencies which outlive the others. The accelerometers pick up the stress waves displacements but the software should realize that these aren't parts of modes because they damp out quickly.

 

[cdemir]

Dear sir;

The problem is the similar to my problem. First of all , I am Research Asist. My working subject is modal analysis of 6x6 vehicle which is form of space frame.
I solved the problem by using finite element.Then took measurement. There is difference between two.

We have Bk 2515 Vib. Analyser. So it has one channel.
I used accel. transducer for measuring. The method is hitting a hammer then frequency spectrum was gained.

So I was confused. What is the problem? Impact hammer method is wrong for a vehicle.

The other point is, each measurement differed the other

What advice do you give me for right measurement?

I can use shaker but how can I make measurement?

(I controlled finite element method and its results million times)
Thanks in advance

 

[Franko]

Who cares if there are stress waves when you strike the structure, if you are mainly interested in several of the lowest natural frequencies. With proper modal tests, it's the frequency content, amplitudes, and relative phases between force and response that must be handled correctly for each direction for a 3-D structure. The strikes and response for a particular direction are evaluated; the correct hammer and tip surface must be used for the frequency range of interest; the force pulse amplitude must be high enough; and the accelerometer must be correctly mounted and sensitive enough to obtain accurate (coherent) data between force and response in the frequency range used for the analysis. Good software has build-in coherence checks. (There could be small structures at high frequencies where impact modal analysis won't work - perhaps holography would be required.) We manufacture turbines and compressors and previously used a HP (Hewlett Packard) System, and now use &quot;Seattle Sound and Vibration&quot; software, with HP FFT analyzers (now Agilent's). I'm sure you can obtain good application guidelines from each company or others; see

http://contact.tm.agilent.com/tmo/press/English/PRTMA100802.html