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: Exact Equation - theory

  1. Jul 18, 2005 #1
    An exact equation has the form

    [tex]M(x,y) + N(x,y) \: y^{\prime} = 0[/tex]


    [tex]M(x,y) = \frac{\partial \psi}{\partial x} (x,y)[/tex]


    [tex]N(x,y) = \frac{\partial \psi}{\partial y} (x,y) \mbox{.}[/tex]

    If [tex]y=\phi (x)[/tex] and [tex]\psi (x,y) = c[/tex], then

    [tex]M(x,y) + N(x,y) \: y^{\prime} = \frac{\partial \psi}{\partial x} + \frac{\partial \psi}{\partial y} \frac{dy}{dx} = \frac{d}{dx} \psi \left[ x, \phi (x) \right] = 0 \mbox{.}[/tex]

    I can't understand this:

    [tex]\frac{\partial \psi}{\partial x} + \frac{\partial \psi}{\partial y} \frac{dy}{dx} = \frac{d}{dx} \psi \left[ x, \phi (x) \right]\mbox{.}[/tex]

    Any help is highly appreciated.
  2. jcsd
  3. Jul 18, 2005 #2


    User Avatar
    Science Advisor
    Homework Helper

    From the definition of an "exact differential" or "total differential" for a function f(x,y) whose integral is path independent

    [tex] df = \frac{\partial f}{\partial x}dx + \frac{\partial f}{\partial y}dy [/tex]


    If you divide by dx you have

    [tex] \frac{df}{dx} = \frac{\partial f}{\partial x} + \frac{\partial f}{\partial y}\frac{dy}{dx} [/tex]

    and since you have y as a function of x, there is a derivative of y wrt x
  4. Jul 18, 2005 #3

    James R

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    [tex]\frac{\partial \psi}{\partial x} + \frac{\partial \psi}{\partial y} \frac{dy}{dx} = \frac{d}{dx} \psi \left[ x, \phi (x) \right]\mbox{.}[/tex]

    Well, [itex]\psi[/itex] is a function of x and y, so

    [tex]\frac{d\psi}{dx} = \frac{\partial \psi}{\partial x} + \frac{\partial \psi}{\partial y}\frac{\partial y}{\partial x}[/tex]

    Now put [itex]y=\phi (x)[/itex], and that's all there is to it.
  5. Jul 19, 2005 #4
    Thank you, guys! Thus, we have

    [tex]M(x,y) + N(x,y) \: y^{\prime} = 0[/tex]

    [tex]d\psi = M(x,y) \: dx + N(x,y) \: dy = 0[/tex]

    [tex]\frac{d\psi}{dx} = M(x,y) \: + N(x,y) \: \frac{dy}{dx} = 0 [/tex]

    [tex]\frac{d\psi}{dx} = \frac{\partial \psi}{\partial x} + \frac{\partial \psi}{\partial y} \frac{dy}{dx} = 0 \mbox{.}[/tex]
  6. Jul 19, 2005 #5
    I'm not exactly sure of how pragmatic it is to convert an ordinary differential equation into a partial differential equation. Maybe I miss something?
  7. Jul 19, 2005 #6


    User Avatar
    Science Advisor

    No one changed an ordinary differential equation into a partial differential equation. They just used partial derivatives to show show that the sum of two differentials can, in some circumstances, be written as a single differential- which makes the problem very easy.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook