1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Finding the frequency of an arbitrary, time-varying load

  1. Jan 28, 2013 #1
    Hi everyone,

    So I have time-varying force data. The force acts as a concentrated load. The force does not vary periodically/harmonically, it just has a general variation with time. My main question is: How do I determine the corresponding frequency of this force data?

    Also, if I'm using MATLAB and I have a data vector of this force's time variation, how do I use MATLAB to determine the corresponding frequency data? Do I use the fft() function?

    Any help/insight would be appreciated! Let me know if you need any more details and I will happily provide them.

  2. jcsd
  3. Jan 28, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    If it is not periodic, it does not have a single frequency.
    That looks like a Fourier transformation: Good. It will show you the main frequency components of your signal.
  4. Jan 28, 2013 #3
    Thanks for the helpful reply, I appreciate it.

    What exactly does fft() give me? For instance, say I define my force vector as V=[2 6 2 9 10 3 . . . .] then plug in fft(V) in MATLAB. What is the physical significance of the resulting numbers? Are they the frequency of the data of the vector?

    Basically my goal is to first plot my force as a function of time from the given data, F(t). I have this data so its very easy. Then I want to plot the force as a function of its frequency, F(f) using a Fourier transform. How do I do this?

    Any help/advice would be greatly appreciated. Let me know if you need my to provide more info.
  5. Jan 29, 2013 #4


    User Avatar
    2017 Award

    Staff: Mentor

    Check the documentation?
    The mathematical interpretation of a Fourier transformation (FT): If you see a peak at some value x, your data has an oscillating signal with a frequency of x. If you see multiple peaks (or even some broad distribution), you have multiple different oscillations in your signal at the same time (or chaos).

    Should be somewhere in the manual.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook