1. PF Contest - Win "Conquering the Physics GRE" book! Click Here to Enter
    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!

Error propagation

  1. Oct 24, 2009 #1
    1. The problem statement, all variables and given/known data

    For my lab work, I have created a theoretical model that goes something like:

    T = \sqrt{\frac{ks^2}{x \sin \theta \cos^2\theta}}

    where k is a constant, and the variables to be differentiated are x, theta and s. How do I find the error of T? I can find the errors of x and s (dx/x and ds/s) from experiment. And of course

    [tex]d\theta / \theta[/tex] can also be found from the instrument.

    3. The attempt at a solution

    Differentiating the term for theta,

    [tex]\frac{dF}{d\theta} = -\frac{1}{2} \sqrt{\frac{1}{\sin\theta \cos^2 \theta}} \cos\theta (1 - 3\sin^2\theta)[/tex]

    It's easier to differentiate the other two. Anyway, I have:

    [tex]\frac{dT}{T} = \frac{ds}{s} + \frac{1}{2}\frac{dx}{x} + \frac{(3\sin^2\theta - 1)\cos\theta}{2}d\theta[/tex]

    How can I express the last term in a [tex]d\theta / \theta[/tex] form? The only thing I can think of at the moment is using a small angle approximation, but I don't know how to justify that. And also, clearly, if I could use that, the small angle approximation for cosine has a square term.
  2. jcsd
  3. Oct 24, 2009 #2
    For a function of multiple variables, [tex]T=f(x, s, \theta)[/tex], the error in the function, [tex]\Delta T=\sqrt{(\frac{\partial T}{\partial x}\cdot \Delta x)^2+(\frac{\partial T}{\partial s}\cdot \Delta s)^2+(\frac{\partial T}{\partial \theta}\cdot \Delta \theta)^2}[/tex]

    I suggest that you stop using the lowercase [tex]d[/tex] for the errors, because you're bound to get it mixed up with the derivatives you're taking.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook