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!

Just a quick one, Have I got my Mass/Spring/Damper model correct?

  1. Oct 20, 2011 #1
    1. The problem statement, all variables and given/known data

    Model the mass/spring/damper sdof system.

    2. Relevant equations

    Below left is a diagram of the system, and the attempt at the solution is obvious the right hand side one.

    3. The attempt at a solution

    Please see attached pdf.

    Thanks for any help!
     

    Attached Files:

  2. jcsd
  3. Oct 21, 2011 #2
    Here's an image of it instead of the PDF, hopefully get a reply now:

    e27b8f33.jpg
     
  4. Oct 21, 2011 #3
    Although I think y=YoSinwt ?
     
  5. Oct 22, 2011 #4

    rude man

    User Avatar
    Homework Helper
    Gold Member

    That looks right if the function of the 'scotch yoke' and motor is to impart a linear excitation y(t). I never heard of a 'scotch yoke' before.

    Are the 'bushes' bushings?
     
  6. Oct 22, 2011 #5
    Yeah the scotch yoke converts the radial motion of the motor into linear motion. And the bushes are just like dry bearings on the rod, which connects the yoke to the spring, that restrict the system to one degree of freedom.

    Thanks for your help :)
     
  7. Oct 22, 2011 #6

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Looks OK except why did you reverse the positions of the mass and the damper?

    What are you planning to do with your model, if anything?
     
  8. Oct 22, 2011 #7
    Just because it was easier to draw. our lecturer said that the way it's hooked up it acts as if it were connected to ground. So I drew it like that.

    It's for my assignment, we took measurements of time and acceleration from the free vibration response of the system and then used the electric motor to take forced response measurements. There's an accelerometer on the mass and the top of the spring so we can compare the input and output measurements. If that makes sense??

    Basically we have to predict the frequency response of the system and compare our predictions to the actual test results. Im not 100% sure how to though :( I've got some of the way there, got the damping rato, natural and damped frequency of the system, and I think I have the parameter values too, these i got from the free vibration test results.

    Got to make equations for magnification factor vs frequency ratio and then phase lag vs frequency ratio to show the frequency response now.. I think I have found them from my notes and all I have to do is change the values of " r " for frequency ratio and plot the answers ? That sound about right?
     
  9. Oct 22, 2011 #8

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Are you just looking for acceleration; what about displacement and velocity? Displacement you can possibly get from motor position (angle position indicator). For velocity, I don't know. I wouldn't want to try to time-integrate the accel output. I kind of gather you're just looking for acceleration of the mass.

    The theoretical part is easier. You'll have to write the (1st order linear, constant-coefficient) differential equation for the system, with the forced excitation as the forcing function. That's very straight-forward if you've had diff. eq.'s. If not, find someone who did!

    By all means use those notes. Otherwise, all the info you need is in the solution of the diff. eq. but it might be burdensome to plot.
     
  10. Oct 23, 2011 #9
    Yeah we just took the voltages from the accelerometers, then I used them in the logarithmic decay method to find damping ratio etc for the first part..So I don't think we need velocity and displacement?

    What I've got to do is use theory to predict the amplitude ratio and phase lag of the mass as a function of the forcing frequency ratio. Then compre the results with the test data. It also says for this test the system response is to be the ratio of, the output displacement of the mass, and the input displacement to the spring, and the phase lag will be considered too. I can't quite get my head around what that means though.

    (btw is forcing frequency ratio ω, or is that just forcing frequency?)


    Also to write a "1st order linear, constant-coefficient differential equation for the system, with the forced excitation as the forcing function", how is the forced excitation, or anything else for that matter, made the forcing function, or function of the equation? Is it that the equation will be something like x="loads of things multiplied and divided by the thing you want the function to be"?

    I'm very new to this so I may not be making much sense or being vague so I do apologise! I appreciate the help :) Any further reading you could recommend to help with this would be good!
     
  11. Oct 23, 2011 #10

    rude man

    User Avatar
    Homework Helper
    Gold Member

    ω
    I noticed you said the system rsponse is considered to be the displacement of the mass compared to the displacement of the forcing function (scotch yoke). I was wondering how you propose to determine the instantaneous position of the mass, which is the output they apparently want. The accel will only give you the 2nd time derivative of displacement. This should be your first concern here.

    You mentioned frequency and phase response graphs before. With those graphs and expressions you don't need to know how to solve the differential equation, it's already done for you. All you do is to compare your data with the graphs and expressions for amplitude and phase.

    Of course, what will happen is that as your forcing function frequency ω gets close to the natural frequency ω0 of your system, the response will peak and the phase will be close to zero. Your charts should indicate that, as should your expressions. BTW ω is frequency, actually radian frequency which is frequency times 2∏, or ω = 2∏. (For example, your 115V outlets put out f = 60 Hz which is 377 rad/s. The frequency ratio you'll be using as your independnt variable is ω/ω0.
     
  12. Oct 25, 2011 #11
    I think because we had an accelerometer on the input and the output, i.e the yoke and the mass, the voltage output of these is the same ratio as if we had actual displacement values. So the comparison between the two can still me made.

    I have plotted my frequency response graphs, as expected the experimental results differ slightly from the theory. out of 4 tests i would say two of them are close enough to seem like the loss in the system due to friction etc could account for the difference. The other two however, both being around 0.9 frequency ratio (close to resonance) are massively different. The theory MF values are 3.9 and 3.66, and the experiment values are 1.3, and 1.4 for the same frequency ratio values.

    Would you say that this is normal? My initial thoughts are either something is wrong here or there is something else happening. I watched a lecture online that talked about Beat response, am i right in thinking this could be the reason? It seems possible but i need to read up on it more and see if it applies to the situation i have here with the results..
     
  13. Oct 25, 2011 #12

    rude man

    User Avatar
    Homework Helper
    Gold Member

    As long as you stick to theoretical vs. empirical (experimental) displacement RATIOS AT A GIVEN FREQUENCY, you're right, accel ratio is same as diplacement ratio.

    However, you seem to have graphed displacement vs. frequency, and that assumption is not correct.

    Diplacement x = x0*sin(wt), acceleration d2x/dt2 = -w2*x0*sin(wt) = -w2*x.

    'Beats' do not exist. The accel output always puts out the forcing frequency, and only that frequency.

    I need to know (1) exactly what you graphed and compared with, and (2) what frequencies you ran. For eaxmple, the accel output: it's a sine wave, did you accurately determine the peak-to peak voltage, did you use a voltmeter or even a lock-in amplifier, or demodulator?

    Your accel data should have been way more than displacement, relatively speaking, at the higher frequencies vs. at the lower.
     
    Last edited: Oct 25, 2011
  14. Oct 25, 2011 #13
    1)

    For the theory plot if you take a look at this pdf i found:

    http://www.colorado.edu/engineering/CAS/courses.d/Structures.d/ASEN3112.Lect21.d/ASEN3112.Lect21.pdf [Broken]

    ...on page 21-5, take a look at the equations in the box labeled (21.7). I have used the those to plot the theory by adding different values of "r" from 0.1 - 2.0. Accept i have in my notes that for the phase it is inverse tan so i did that.

    For comparing these theory plots, i used the experimental data. This is 4 sets of data, each a set of two times and 4 voltages for the input peak and the output peak that follows it, (the four voltages are peak to peak) and also we had another time for the next input peak. From these i worked out the time period, lagging time of the output, amplitude input and output. then with that data i worked out the MF, phase lag and forcing frequency as follows:

    MF=Amplitude output / amplitude input

    phase angle=( lagging time of output to the input / forcing time period ) * 360

    forcing frequency = (1 / forcing time period) * 2pie


    2.)

    We used an oscilloscope and recorded a small amount of the sine waves, then tracked along the display to the peaks and recorded values for the input and output sine waves as above.

    the frequencies we ran at are roughly 28rads, 29rads, 44rads, 63rads. I don't know why the lecturer did 28 and 29 rads so close together though ?



    Edit:

    Just for reference, my graphs look pretty similar to figure 21.2 on that pdf for the lines of damping ratio=0.1..... my damping ratio is 0.125
     
    Last edited by a moderator: May 5, 2017
  15. Oct 25, 2011 #14

    rude man

    User Avatar
    Homework Helper
    Gold Member

    What sort of device is attached to the motor or whatever to measure displacement?

    Was your 'scope input set to dc or ac? Scopes will sharply attenuate frequencies below a certain value if the input is ac-coupled, which could help explain your data.

    What frequency gave you your highest MF?

    Finally, give me your input and output peak-to-peak voltage readings for the lowest and the highest excitation frequencies you used.
     
    Last edited: Oct 25, 2011
  16. Oct 26, 2011 #15
    Theres nothing attached to the motor to measure displacement, just an accelerometer to measure acceleration.

    Im not sure if the scope input was ac or dc? Is 'scope input' a setting on the oscilloscope? I will ask my lecturer but I'm not sure what scope input is so not sure what to ask.

    In the theory i got my biggest MF at 0.98 frequency ratio, i thought that was supposed to be at frequency ratio of 1?

    the peak to peak input and output for lowest and highest are as follows:

    Lowest excitation was 29rads:
    Input peak to peak = 97.74mV
    output peak to peak = 147.88mV

    Highest excitation was 63rads:
    Input peak to peak = 448.15mV
    Output peak to peak = 95.1mV

    the two lowest excitations are the ones that are significantly lower that theory states they should be.
     
  17. Oct 26, 2011 #16

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Ah, so now I know. You're measuring input acceleration wheras I thought the sensor measuring input was measuring displacement. I see you had mentioned this already, sorry I didn't pick up on it sooner.

    So now everything's fine. Your ratio is output accel/input accel which will be the same as output displacement/input displacement. So you can compare your results with the charrts you have.

    You still didn't give me the frequency at which MF peaked, just the ratio.

    The fact that you got a lower ratio than you expected at the lowest frequencies seems to corroborate my suspicion that your 'scope was ac-coupled. Yes, ac vs. dc coupling is a selectable input option on every 'scope I've ever seen, though some real low-end ones might not have that option. Check with your lab instructor. Check with me as to what to dowith your data if this turns out to have been the problem if you want.

    The ratio of 0.98 is probably right-on. When damping is added to a sytem like yours, the frequency of max MF is lower than the natural frequency, which is √(k/m).

    See http://en.wikipedia.org/wiki/Harmonic_oscillator for a good discussion.
     
    Last edited: Oct 26, 2011
  18. Oct 26, 2011 #17
    Oh yeah sorry the frequency of the max MF was 31.17rads

    Ok I'll send him an email, I should hear back tomorrow he's usually pretty good with replying. I haven't got time to read the wiki article till tomorrow but thank you very much!

    By the way, is it really ok to discuss coursework on here? If it came up in the plagiarism checks they do I'm assuming it's just the same as discussing the problem with anyone in person or getting the answer from a text book so it shouldn't be a problem?
     
  19. Oct 26, 2011 #18

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Obviously, that depends on the view taken by your school.

    This site is supposed to give students a hint, never an off-the-top solution. I myself don't even bother replying to anyone who hasn't evinced significant effort to solve the problem beforehand.

    I believe this site is well known so if your school has a problem with it I would think they would have warned you. Obviously, you would not want to cheat on an in-class exam with your android using it!
     
  20. Nov 9, 2011 #19
    Hi,

    I have finally sorted the lower values out! Turns out if i use different method for working them out they come out about right.

    I have however got two values on the phase lag vs frequency response graph that are odd. Two of the results are close to the theory plot at around r=0.8 to r=0.9, the other two are about 10 degrees lower than the theory. So the phase lag is less than predicted for the two tests where the system was ran at higher speeds.

    Would this be something expected or is something wrong here? My thoughts are that the friction in the system is preventing the phase change making a full 180 degree swap. Am I on the right track there?

    ScreenShot2011-11-08at1915.jpg


    Also, im so close to getting the equation of motion right for the system. Could you hint me in the right direction?

    EoMOwenJeffs.jpg
    IMG_0644.png
     
  21. Nov 9, 2011 #20

    rude man

    User Avatar
    Homework Helper
    Gold Member

    I forgot - is the mass dangling against gravity or on a frictionless horizontal plane? Since you don't have g in your equation I'll assume it's on a horizontal table or similar.

    Now, as to your equation - I don't get the same thing, so could you go thru the steps you took to arrive at it?

    In parti cular, where did the cwy0cos(wt) term come from?
     
    Last edited: Nov 9, 2011
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook