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!

Homework Help: Theoretical model for damped harmonic oscillation.

  1. Feb 19, 2017 #1
    1. The problem statement, all variables and given/known data
    Hello all,

    I have a question regarding the damping constant for a model of a vertically oscillating mass on a spring. I have read through one or two similar questions on this site but I think I can manage to be a little more specific about what I'm asking.

    I am in a physics lab course and have collected data from a motion detector for a mass on a spring by itself, and two sets of data for which wide, flat objects were affixed to the mass (with some metal disks removed from the mass to offset the additional mass from the 'damper').

    I am asked to create a theoretical model for the equation of motion and compare it to our collected data. The issue I'm having is when I try to predict the damping constant, b. I am hoping to find an expression for drag force which is linear with respect to velocity; the constant be needs to be such that multiplying it by v gives units of force.

    The model of drag force I have learned previously is dependent on (v^2)/2 and so throws off my model. I have looked into Stoke's equation since our system never reaches very large velocity, but treatments of Stoke's seem to specify that it applies only small spherical objects moving slowly enough that the airflow relative to the object is completely laminar.

    So my question is this; is there a way predict b (even with somewhat low accuracy) from the area and possibly an average velocity which gives be in units of kg/s? or alternatively, is there a way to take the fairly simple drag force equation [(CA{rho}v^2)/2] and make it dependent on v instead of v^2?

    Or any other ideas for how I might make a reasonable guess at b for the sake of creating a theoretical model to compare my data to. I am not versed in fluid dynamics beyond the basics like Bernouli's and the equation of continuity for laminar flow, but I will make every attempt to learn the necessary math and physics if someone could point me in the right direction.

    2. Relevant Equations; unknown (or stated above)

    3. Attempt at the solution;

    Algebraic manipulation and seemingly endless forum searches.
  2. jcsd
  3. Feb 19, 2017 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    A theoretical model does not necessarily have to predict the constants. It might be enough to predict the shape of the relationship (linear or quadratic) and obtain the parameters experimentally.
    Why do you think it should be linear with velocity? Have you calculated the Reynolds number?
  4. Feb 20, 2017 #3
    @haruspex: I believe the constant b should be linear with velocity because the model for damped harmonic oscillation is a solution to the equation of motion for a harmonic oscillator, where the force responsible for damping is described as F = -bv. If the constant b then changes with velocity it introduces a whole new set of problems lab report (which I'm not sure I have time to learn and write about before this is due.)

    I have spoken to teachers at my university since posting this question who have suggested I model the drag force as if one of the velocity terms is constant, based on the approximate average of the velocity of the system. I am going to move forward with this strategy, but if anyone would like to discuss it here, please feel free!

    I have been given all the help I need, I believe. Thanks for your reply.
  5. Feb 20, 2017 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Yes, that is the classical model referred to as a damped harmonic oscillator, but I see no reason why the physical set up you describe should be dominated by viscous forces. You wrote that you are asked to create a theoretical model for the actual experiment. Quadratic seems more likely.
    And, yes, quadratic is far harder to deal with.
    Anyway, following the advice of your teachers seems the best move.

    Edit: more thoughts...
    A theoretical model can just be a differential equation. It does not have to be solvable analytically. If you have the time, I suggest setting up a quadratic model in software (just a spreadsheet would do) and see which can be tuned to give the better fit to the data.
    Last edited: Feb 20, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted