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!

I got other method of solving

  1. May 30, 2009 #1
    Vs(t)=V_0 U(t)
    (Vc)' + \frac{1}{RC}Vc=\frac{1}{RC}Vc=\frac{1}{RC}V_0
    (Vc)' + \frac{1}{RC}Vc(t)=\frac{1}{RC}V_0
    from her i cant under anything regarding the term of use:
    homogeneous solution is:
    we guess a solution from the form of
    and substitute into the homogeneous equation:
    \int Ae^{st} +\frac{1}{rc}Ae^{st}=0

    these are only the first two steps but i cant understand why are they doing that
    the youtube solution differs alot

  2. jcsd
  3. May 30, 2009 #2


    User Avatar
    Science Advisor

    I don't understand this. Did you mean to have those two "=" or is this a typo?

    Okay, is this what you meant above?

    Well, that's the "associated homogeneous equation", not yet a "solution".
    You can rewrite it as [itex]dV_{ch}/V_{ch}= (-1/RC)dt[/itex] and integrate that.
    [itex]ln(V_{ch})= -t/RC+ K[/itex] and taking exponentials of both sides, [itex]V_{ch}= K_1e^{t/RC}[/itex] where [itex]K_1= e^K[/itex].

    Why an integral? [itex]V_{ch}'[/itex] is the derivative
    [tex](Ae^{st})'+ \frac{1}{rc}Ae^{st}= sAe^{st}+ \frac{1}{rc}Ae^{st}= 0[/tex]
    so the exponentials cancel leaving s+ 1/rc= 0. s= -1/rc and the solution becomes [itex]Ae^{t/rc} just as before. That method is more often used with higher order differential equations where you cannot integrate as I did above.

    You can justify that method by arguing that for y(x) something like Ay"+ By'+ Cy= 0, a "linear equation with constant coefficients", in order that y and its derivatives cancel, to give 0, y' and y" must be the same "kind" of function as y. Exponentials do that nicely: the derivative of [itex]e^{ax}[/itex] is [itex]ae^{ax}[/itex], the same exponential multiplied by a.

    But you should be aware this is not based on any idea that a solution MUST be an exponential! For example, The equation y"= 0 has general solution y= Ax+ B and y"+ y= 0 has general solution y= A cos x+ B sin x. Since those solutions are indirectly related to exponentials, "trying" [itex]e^{sx}[/itex] can still lead you to them.
    Last edited by a moderator: May 30, 2009
  4. May 30, 2009 #3
    U(t) is called the shock function
    it equals 1 in this case
    Vs=vo*U(t) (Vs the voltage of the source)

    does this help ??
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: I got other method of solving
  1. I got one problem (Replies: 3)

  2. I got 3 questions! (Replies: 4)