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: Find a complex number from differential equations.

  1. Jan 6, 2010 #1
    1. The problem statement, all variables and given/known data
    Find complex number [tex]\lambda[/tex] such that e[tex]\lambda[/tex]t solves
    [tex]\frac{d^{2}y}{dt^{2}}[/tex] + 4[tex]\frac{dy}{dt}[/tex] + 5y = 0

    Express this solution in the form eat(cos(bt) + i sin(bt))
    2. Relevant equations

    3. The attempt at a solution
    So the first part is fine, using [tex]\lambda[/tex]2 + 4[tex]\lambda[/tex] + 5 = 0 to get values of [tex]\lambda[/tex] at -2[tex]\pm[/tex]i. From here, I've been taught to use:

    y = Ae[tex]\lambda[/tex]1t + Be[tex]\lambda[/tex]2t but this time, it doesn't help get to the required form.

    Using what I've mention, I can get to A=-2 but finding B seems to be a mystery. Any help greatly appreciated!
  2. jcsd
  3. Jan 6, 2010 #2
    Are you aware of Euler's identity?
    e^{ix} = \cos x + i \sin x
  4. Jan 6, 2010 #3
    I am. Converting to (and from) polar form. My attempt went as far as:

    y = [tex]Ce^{-2t+it} + De^{-2t-it}[/tex]
    y = [tex]e^{-2t}(Ce^{it}+De^{-it}[/tex])

    That's how I got to A=-2 but from here:

    y =[tex]e^{-2t}[/tex] (C(cos(it) + i sin(it)) + D(cos(-it) + i sin(it)))

    and from there, getting a value for the B in question is eluding me.

    Note: I've changed the constants in the equation so they don't clash with the required values.
  5. Jan 6, 2010 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    I think the problem is just asking you to express each of the solutions in the form given separately, before you get to the step of writing down with the general solution.
  6. Jan 6, 2010 #5
    it was just the question was phrased:
    "Find complex number [tex]\lambda[/tex] such that e[tex]\lambda[/tex]t is a solution to the equation (see above). Express this solution in the form (see above) and give the values of real numbers A and B"
    Just that seemed like there was one :/
  7. Jan 6, 2010 #6
    y = e^{-2t} (C(\cos(it) + i \sin(it)) + D(\cos(-it) + i \sin(-it)))

    Note the following:

    \cos -x = \cos x
    \sin -x = - sin x


    y = e^{-2t}[ (C+D) \cos (it) + (C -D) \sin(it)]

    We can let C+D and C-D be constant by themselves, say K and M
    y = e^{-2t}( K \cos (it) + M \sin(it))
    This is all just to make it pretty and for future problems. For this particular problem, all that you needed to know was that sine was odd and cosine even.
    Thus, the b value is, of course, one.
  8. Jan 6, 2010 #7
    i see! right, so maybe a quick sketch of sine and cosine graphs would have cleared this up for me! I'll remember that in future! Thanks a load for all the help!
  9. Jan 7, 2010 #8


    User Avatar
    Science Advisor

    No! [itex]e^{it}= cos(t)+ i sin(t)[/itex], not [itex]e^{it}= cos(it)+ i sin(it)[/itex]

    [tex]y= e^{-2t}(C e^{it}+ De^{-it})= e^{-2t}(Ccos(t)+ Ci sin(t)+ D cos(-t)+ Di sin(-t))[/tex]
    Now, cos(-t)= cos(t) and sin(-t)= -sin(t) so this is
    [tex]y= e^{-2t}(C cos(t)+ Ci sin(t)+ D cos(t)- Di sin(t))= e^{-2t}((C+D)cos(t)+ (C-D)i sin(t))[/tex]

  10. Jan 7, 2010 #9
    Of course. Sorry, slip of the keyboard ;)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook