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: Diverge? homework help

  1. Dec 18, 2005 #1
    :cry: I am in doubt about convergence of the folowing improper integration:
    \ x * sin(kx) dx; (k is real)

    sorry about my sign of integration! I promisse that I will learn how to post correctly!

    I used the intuition:
    sin is periodic and bounded and x is unbounded. In my intuition this integration diverges. Bur I need rational answer, intuition in mathematics is something of nothing.
    Last edited: Dec 18, 2005
  2. jcsd
  3. Dec 18, 2005 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What is sen?

    Have you tried any theorems that let you prove an integral diverges?
  4. Dec 18, 2005 #3


    User Avatar
    Gold Member

    All you need to do to prove that the integral diverges is appeal to the definition of the improper integral:
    [tex]\int_{0}^{\infty}x\sin{x}dx \equiv\lim_{b\rightarrow\infty}\int_{0}^{b}x\sin{x}dx[/tex]
    You can evaluate the integral in terms of b using integration by parts and then show that the limit as b approaches infinity does not exist.
  5. Dec 18, 2005 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Intuition is good; but you're right, you need a proof in the end.

    However, many definitions and theorems in mathematics are designed to capture our intuition. In particular,

    This is good intuition, and I'm sure that there's a theorem that says almost exactly this, but I can't remember what it is off hand.

    But, if you can't find the theorem, you could always try to prove it:

    Theorem Let [itex]f(x)[/itex] be a nonzero periodic function with period [itex]L[/itex], and let [itex]g(x)[/itex] be an unbounded function. Then, the integral

    [tex]\int_{0}^{+\infty} f(x) g(x) \, dx[/tex]


    Actually, the theorem I just stated is false. For example, you could let [itex]f(x)[/itex] be the function:

    f(x) := \left\{
    \begin{array}{l l}
    0 \quad & \lfloor x \rfloor \mbox{\ is even} \\
    1 \quad & \lfloor x \rfloor \mbox{\ is odd}

    And let [itex]g(x)[/itex] be the function [itex]g(x) := f(x - 1) \cdot x[/itex]

    But a suitable modification to the "theorem" I stated (e.g. it would suffice to make [itex]g(x)[/itex] a strictly positive function) would be true. See if you can prove it!

    (P.S. what level math are you in?)

    LeonhardEuler's approach would work too, so you could do the problem that way. However, it would be much nicer to have a theorem you could use for problems of this type!

    (P.S. if you click on any of our math formulae, you will see what you need to type to make that formula)
    Last edited: Dec 18, 2005
  6. Dec 18, 2005 #5
    Thank you for your great help!
    I am just a beginner and very lazy!
    I think that Lazy is mortal sin for everyone but in special for a mathematician.

    The TEX used in this forum is the standart?

    I plan to start to learn TEX and this site will help me so much in my learning!

    Thank you so much!
  7. Dec 27, 2005 #6


    User Avatar

    Unfortunately, this integral has no analytical solution. He can try to develop sin x in series, integrate the series and see if it converges.
  8. Dec 27, 2005 #7


    User Avatar
    Science Advisor
    Homework Helper

    Yes it does. The integral evaluates to [itex]\sin b - b \cos b[/itex].
  9. Dec 28, 2005 #8


    User Avatar
    Science Advisor

    Of course it does: integrate by parts.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook