Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Experimental Modal Analysis

  1. Feb 11, 2008 #1
    Hello, I need some help with setup/equipment used for a study to determine the natural frequencies on some small, lightweight samples (12in x 1.5in x .12 in.) in a catilever beam setup. The samples are made of carbon fiber and kevlar so they are fairly flexible to begin with. I am working on an undergrad thesis and I am very new to this type of testing.

    I am currently trying to use the equipment that is already available to me: Impact hammer with signal conditioner and oscilloscope. I am wondering if this setup will yield accurate results or do I need something different for such a lightweight application.

    There is also a shaker available to me with a computer and software but I opted to try the impact hammer first because the shaker, rated for 300lbs. of force seems much to large for these samples.

    Any suggestions would be appreciated.

  2. jcsd
  3. Feb 11, 2008 #2


    User Avatar
    Science Advisor
    Gold Member

    It's probably a good idea to first do some hand calcs to see what the first couple of vibrational modes are, to see what frequencies you will be working at. That will help you decide what setup will be best suited for what you're trying to do.
  4. Feb 11, 2008 #3
    Like i said, I have never done any work in vibration and have no idea what types of hand calculations I would need. I will attempt to do some research on that.

    Maybe you can answer this question, what is the differenct between a FFT spectrum, and a FRF function? I am able to generate a FFT spectrum on the oscilloscope but is it possible to determine the mode frequencies from that?
  5. Feb 11, 2008 #4


    User Avatar
    Science Advisor

    If it is really flexible, you will probably not get good results with the hammer. The flexing will attenuate the higher order modes. You really need to do it on a shaker.

    The FRF is the result of taking an FFT of the time domain data. It is frequently interchanged by calling the FRF the FFT. If you can do the FFT on the scope (I would assume you have some kind of hold or peak hold function) then you will see the FRF. The large peaks on the FRF will be your modes.

    Let me see if I can dig up some basic modal testing literature for you.
  6. Feb 14, 2008 #5
    I've been trying to decipher some textbooks on vibration. The problem is that I can't find a straightforward method for how to perform these hand calculations. Mainly I don't know what type of assumptions to make for my system. For example, is it free or forced vibration?, can I assume it's harmonic?, how many degrees of freedom are there?, And is there damping I need to account for, and what type of damping is it?

    Having never done any problems like this, I'm completely lost.

    Please Help!!
  7. Feb 18, 2008 #6
    wow it's so amazing to see someone in exactly the same situation as me! i'm currently trying to find the natural frequencies of a beaker (well ultimately a beer keg) and it is NOT working out for me. My supervisor wants me to use the circle fit method and i've been researching it for ages and still haven't a clue how to actually do it. Sorry i can't help but thought you'd like to know you're not alone!
  8. Mar 1, 2008 #7
    The FRF is calculated from the time domain data by calculating the cross and autopower spectral densities of the input and output time domain data (this involves taking the FFT of the time domain data). Assuming that x and y are the input and output signals respectively the H1 and H2 FRF estimators are calculated as: H1 = Sxy/Syy and H2 = Sxx/Syx. S denotes the auto (Sxx and Syy) and cross (Sxy and Syx) power spectral density of the input and output signals.

    These peaks also contain contributions of all other modes in your structure. How much this contribution is depends on the structure under consideration. Generally, if the peaks in the FRF are not too close then the actual mode shape can be estimated reasonably well directly from the amplitude and phase of the FRF. The difficulty is determining when peaks are too close.

    Some good online information on experimental modal analysis and vibration can be found at: http://www.sdrl.uc.edu/academic-course-info [Broken] Look at the vibrations I, II and III courses. I think the vibrations III course is about experimental modal analysis.
    Last edited by a moderator: May 3, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook