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!

Implicitly Deifned Parametrization

  1. Apr 23, 2006 #1
    Implicitly Defined Parametrization

    I'm having difficulties with the following question, and having checked through my working several times I just can't find a problem...problem is, so far in the book implicit and parametric differentiation have been covered independently of each other and this question has just been thrown into the mix.

    Find the slope when t = 0:
    [tex] x + 2x^{\frac {3}{2}} = t^2 + t[/tex], [tex] y \sqrt{t+1} + 2t \sqrt{y} = 4 [/tex]
    I can't see any other way to work with this other than to differentiate both equations w.r.t.t, find dy/dx by dividing dy/dt by dx/dt, and then plug in my numbers at the end.

    dx/dt: [tex] \frac{dx}{dt} +3 \sqrt{x} \frac{dx}{dt} = 2t + 1[/tex]

    [tex] \frac{dx}{dt} =\frac{2t + 1}{3 \sqrt{x}+1}[/tex]

    dy/dt: [tex] \sqrt{t+1} \frac{dy}{dt} + \frac {y}{2\sqrt{t+1}} + \frac{t}{\sqrt{y}}\frac{dy}{dt}+ 2\sqrt{y} = 0[/tex]
    just gonna assign dummy variables:[tex] a = \sqrt{y}, b = \sqrt{t+1} [/tex] whilst I rearrange things...
    [tex] \frac{dy}{dt}(b + \frac{t}{a}) = -(\frac {y}{2b} +2a)[/tex]

    [tex] \frac{dy}{dt}(\frac{ba +t}{a}) = -(\frac {y +4ab}{2b})[/tex]

    [tex] \frac{dy}{dt} = -(\frac {a(y +4ab)}{2b(ba+t)}) = -(\frac {\sqrt{y}(y + 4\sqrt{y}\sqrt{t+1})}{2\sqrt{t+1}(\sqrt{y}\sqrt{t+1}+t)})[/tex]

    dy/dx =[tex] -\frac {(3 \sqrt{x}+1)(\sqrt{y}(y + 4\sqrt{y}\sqrt{t+1}))} {(2t + 1)(2\sqrt{t+1}(\sqrt{y}\sqrt{t+1}+t))}[/tex]

    By using t = 0 and finding x and y in the original equations I get x = 1/4 and y = 4
    [tex] -\frac{(\frac{3}{2}+1)(8 + 16)}{4} = -15 [/tex]

    The answer I'm looking for however is -6. Is there some screw up with my working somewhere or am I using the wrong method?
    I cannot graph any similar (but simpler) curves with Maxima to find the correct method neither because there doesn't seem to be a way to graph implicit expressions...nor will it differentiate one
     
    Last edited: Apr 23, 2006
  2. jcsd
  3. Apr 23, 2006 #2

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I did not check all the steps (the parenthesis don't match in the denominator in yoru final dy/dx) but how did you get x=1/4? When t =0, x=0 !
     
    Last edited: Apr 23, 2006
  4. Apr 23, 2006 #3
    Fixed the parenthisis :smile:
    By setting t =0 in:[tex] x + 2x^{\frac {3}{2}} = t^2 + t[/tex]
    I get :[tex] x + 2x^{\frac {3}{2}} = 0[/tex]

    [tex] x^{\frac {1}{2}} = -\frac{1}{2}[/tex]
    x = 1/4

    That response has just made me see how -6 is possible tho :smile:
     
    Last edited: Apr 23, 2006
  5. Apr 23, 2006 #4

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    You are right in that there are two solutions. The other solution (x=0) gives indeed -6.
    It`s all a question of which sign to take when calculating a square roo. For your answer to work one must use the convention that one must take the negative sign for a square root. I was assuming that the convention was to take the positive root.
     
  6. Apr 23, 2006 #5
    Thankyou very much Nrged for putting my mind at ease and pointing out a solution I kept overlooking...what I did *lost* a solution! :smile:
     
    Last edited: Apr 23, 2006
  7. Apr 23, 2006 #6

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    Actually, no. [itex]x^{\frac{1}{2}}= -\frac{1}{2}[/itex] has no solution since the 1/2 power of any real number is non-negative. The only root os [itex]x+ 2x^{\frac{3}{2}}= 0 is 0.
     
  8. Apr 23, 2006 #7
    Thanks for pointing out where I am wrong HallsofIvy :smile: I won't make this same mistake twice.
     
  9. Apr 23, 2006 #8

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    So if we have x^2 = 4, the solution is only x=2 and cannot be x=-2? (where x is a real) ?
     
  10. Apr 23, 2006 #9
    no because x^2 = 4 does give two roots. +2 and -2.

    -2*-2 = 4

    2*2 = 4
     
  11. Apr 23, 2006 #10

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I know, but HallsofIvy seems to imply the contrary which kind of surprised me . I'd like to understand what he meant (he is very knowledgeable so I must be missing his point entirely)
     
  12. Apr 27, 2006 #11

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I am still confused by that statement: since the 1/2 power of any real number is non-negative. :surprised
     
  13. Apr 27, 2006 #12

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    The equation x2= a has two solutions. They can be written [itex]\sqrt{a}[/itex] and [itex]-\sqrt{a}[/itex]. The reason why we need to write "+" and "-" is because [itex]\sqrt{a}[/itex] is defined to be the positive root! There is a difference between "the roots of the equation x2= a" and "the square root of a". The equation has two roots while the square root function, [itex]\sqrt{a}[/itex], is defined as the positive solution to x2= a.
    a1/2 is also [ b]defined[/b] as [itex]\sqrt{a}[/itex], the positive real root of x2[/sub]= a.
     
  14. Apr 28, 2006 #13

    nrqed

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member


    Ok, that makes sense. Thank you.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Implicitly Deifned Parametrization
Loading...