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: System of differential equations question

  1. Oct 26, 2009 #1
    1. The problem statement, all variables and given/known data
    I have to solve the following system of differential equations, given some initial conditions:

    [tex]mx'' = qBy'[/tex]
    [tex]my'' = -qBx'[/tex]

    3. The attempt at a solution

    Rearranging I get [tex]x'' - \frac{qB}{m}y' = 0[/tex]
    [tex]y'' + \frac{-qB}{m}x' = 0[/tex]

    Computing the operational determinant I get [tex]cD^3 + cD^3[/tex], where C = qB/m. Continuing I apply the operators to x and y to get [tex]2cD^3x = 0[/tex] and [tex]2cD^3y = 0[/tex]. So basically I get [tex]y''' = 0[/tex] and [tex]x''' = 0[/tex]. I'm not sure how this helps me solve the system of differential equations, as with the characteristic equation r^3 = 0 if one root is 0 then the others could be anything...I think I've gone wrong somewhere but cannot see it. Any help would be appreciated.
  2. jcsd
  3. Oct 26, 2009 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Differentiating the first equation gives:

    mx''' = qBy'' and from the second equation y'' = -(qB/m)x'. Substitute that in for y'':

    mx''' = -(q2B2/m)x'

    x''' + c2x' = 0 where c =(q2B2/m2)

    Let w = x' and you have a second order constant coefficient equation in w = x'.
  4. Oct 27, 2009 #3
    If I continue in that way, I find that [tex]x' = C_1 + C_2e^{-Ct}[/tex] and [tex]y' = C_3 + C_4e^{-Ct}[/tex], so x is equal to [tex]C_1t - \frac{C_2}{C}e^{-Ct} + C_3[/tex] and y is equal to [tex]C_4t - \frac{C_5}{C}e^{-Ct} + C_6[/tex]. C is equal to [tex]\frac{q^2B^2}{m^2}.[/tex]

    I now have 6 constants, but I should really only have 2...I also have 4 initial conditions from the problem, namely [tex]x_0 = r_0, y(0) = 0, x'(0) = 0, y'(0) = -\omega r_0 [/tex] where omega = qB/m. Do I plug the solutions for x and y back into the original equations and start equating coefficients and hope that I'll be left with only 2 constants? My goal is to show that the particle moves in a circle with radius [tex]r_0[/tex] but I don't see it yet. Thanks so much for your help!
  5. Oct 27, 2009 #4


    User Avatar
    Homework Helper

    how do you get [itex]x' = C_1 + C_2e^{-ct}[/itex] from [itex] x''' + c^2x' = 0[/itex]?

    [tex]x' = C_1 + C_2e^{-ct}[/tex]
    [tex]x'' = -c.C_2e^{-ct}[/tex]
    [tex]x''' = c^2 C_2e^{-ct}[/tex]

    substituting back into the original equation gives
    [tex]x''' + c^2x' = c^2 C_2e^{-ct} + c^2(C_1 + C_2e^{-ct}) \neq 0[/tex]

    follow LCKurtz suggestion let w = x' and solve
    [itex] w'' + c^2w = 0[/itex]
  6. Oct 27, 2009 #5
    I see it now. I made a mistake setting up my characteristic equation..:redface:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook