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!

Upper envelope of a function

  1. Feb 23, 2010 #1
    1. The problem statement, all variables and given/known data
    find the upper envelope for the family of ballistic curves :

    y = ax - [x^2(a^2+1)]/2

    an upper envelope is a curve y=F(x) such that for each x fixed, F(x) is the maximum of g(a) = ax - [x^2(a^2+1)]/2 for a in R

    2. Relevant equations

    3. The attempt at a solution
    diff g(a) wrt to a and equate to 0 and get a=1/x so F(x) is 1/x?? :(
  2. jcsd
  3. Feb 24, 2010 #2
    You are right to start by finding the maximum of [tex]g_x(a) = ax - \frac12 x^2 (a^2 + 1)[/tex], but you need to prove that the maximum of [tex]g_x[/tex] occurs at the single critical point you find.

    However, you have found the value of [tex]a[/tex] which maximizes [tex]g_x(a) = F(x)[/tex] at that [tex]x[/tex] --- not what you want, which is the value of the function at that point, expressed as a function of [tex]x[/tex]. What further computation is necessary here?
  4. Feb 24, 2010 #3
    Am I right in thinking you then have to sub the a=1/x back into the original y to get,

    y=F(x)=(1/2) x^2

    and mathmathmad your a warwick student right?
  5. Feb 24, 2010 #4
    I do not think so as it is not related with F(x) being maximum of g(a). Btw blackscorpian, I've sent you a pm.
  6. Feb 25, 2010 #5
    is it F(x)=1/2 - x^2/2 instead?

    @blackscorpion : yes... I am. you too?
  7. Feb 25, 2010 #6
    Yes it is
  8. Feb 25, 2010 #7
    Ahhh crap, it is aswell.
    Stupidly cancelled the ones forgettin bout the over 2 part.
    Well thats a mark thrown away, lol
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook