1. Not finding help here? Sign up for a free 30min 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!

Help me please

  1. May 23, 2004 #1
    Help me please!!!!!!!

    Can you help me to solve this:

    (d² e(r)/dr²)+(1/r)*(d e(r)/dr)=0

    There is no initials conditions, please use general form
  2. jcsd
  3. May 23, 2004 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    Convince yourself of the following equality:
  4. May 23, 2004 #3
    Yes, but....

    And, if i want to find the general form of e(r)?
  5. May 23, 2004 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    You get:
    This differential equation can be directly integrated to find the general solution.
    Last edited: May 23, 2004
  6. May 23, 2004 #5
    Maybe this?

    If I understand, you've made:

    If i integrate, i find:

    Wrong or not ?
  7. May 23, 2004 #6

    THANKS!!!!! you save me!!!
  8. May 23, 2004 #7
    I would also do it this way:



    I would then multiply through by r^2:


    I would recognize this as a d.e. of the Euler-Cauchy form:

    [tex]x^2y''+axy' + by=0[/tex]

    In the case of the given equation, a=1 and b=0. The characteristic equation for the Euler-Cauchy is:


    In our case:


    For the case of a real double root in the characteristic equation, the general solution for the Euler-Caucy is given as:

    [tex]y=(A + B\ln x)x^m[/tex]

    So in our case:

    e(r)&=(A + B\ln r)x^0\\
    e(r)&=A + B\ln r

    I guess this solution depends on having the Euler-Cauchy form available to you in your course, which may not be the case.
  9. May 23, 2004 #8


    User Avatar
    Staff Emeritus
    Science Advisor

    Another way to do this problem is let u= e' so that u'= e" and the equation reduces to the separable first order equation u'+ (1/r)u= 0. Then du/u= -dr/r and so
    ln(u)= -ln(r)+ C1 or u= e'= C1/r. Integrating again, e= C1ln|r|+ C2.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?