Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Second, Third (ect) Derivatives, and their relation to the function value

  1. Jul 26, 2012 #1
    Hello, I was pondering acceleration's relation to velocity, when I came across a problem that is more of an issue of differentiation.
    Given our variables

    α=slope of 'a'
    (at1→t2)=Total acceleration from t1 to t2
    t1→t2)=Total slope of 'a' from t1 to t2



    would the two equations above imply that: [itex]α=\frac{{d^2}v}{d{t^2}}[/itex]

    Since I can find (assuming a has no slope) v(t) with: [itex]a_{{t_1}→{t_2}}=\int_{t_1}^{t_2} a\;dt[/itex]

    we can apply this to the slope of 'a' that:

    [tex]α_{{t_1}→{t_2}}=\int_{t_1}^{t_2} α\;dt[/tex]

    would this also imply that the total slope of 'v' with reguard to the slope of 'a' from t1 to t2 be equal to:

    [tex]\int\int_{t_1}^{t_2} α\; dt[/tex]

    I ask this in the differentiation page since it seems this applies generally to calculus. Converting to generall variables (v=y t=x) would:

    [tex]f(x)=\int\int \frac{{d^2}y}{d{x^2}}\;dx[/tex]

    I'm asking to clairify. I also noticed that if you look at this algebriaclly you get (given that [itex]α=\frac{{d^2}y}{d{x^2}}[/itex])
    [tex]α=\frac{y}{x^2}\Rightarrow y=α{x^2}[/tex]
    but that gives a different answer then the integration, unless if I square x (which I only misunderstand because of the fact, that Leibnez's notation squares the 'd' instead of the 'y'). When I do square 'y', I get the same answer as when I integrate.

    so basiclly what im asking is:

    to find the value of f(x), can i integrate all the way from whatever slope of f(x)?

    can i express this algebriaclly by putting y to the power of the [number of slopes], and x to the same power respectivly?

    if [itex]α=\frac{dy}{dx}[/itex] and if [itex]β=\frac{dα}{dx}[/itex], does that automaticly mean that [itex]β=\frac{{d^2}y}{d{x^2}}[/itex]
    sorry if i made any silly asumptions in my math, im new to calculus and math-interpreted-physics, thank you.
  2. jcsd
  3. Jul 27, 2012 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Need to be a bit careful with the limits. Remember that the variable of integration has no meaning (strictly speaking) outside the integral. Using the same symbol in both places is a 'pun', i.e. they refer to different entities by the same name. Clearer is:
    [tex]\int_{u=t_1}^{t_2}\int_{t=t_1}^{u} α\; dt du[/tex]

    Yes, but with the same clarifications.
    You've lost me. Where did [tex]α=\frac{y}{x^2}[/tex] come from?
  4. Jul 27, 2012 #3


    User Avatar
    Science Advisor

    NO, you cannot treat this as if it were algebra.

    It is incorrect to think of those as "squares" at all. They only indicate that this is a second derivative, have nothing to do with multiplying and cannot be cancelled

    I don't know what you mean by "integrate all the way" or "whatever slope"..

    No, you cannnot solve differential equations "algebraically" and those superscripts are NOT powers.

    Yes, that is true but is not "algebraic". Rather, you are just saying that a "derivative of a derivative is a second derivative."

    Last edited by a moderator: Jul 28, 2012
  5. Jul 28, 2012 #4
    ok no algebra, can I still do algebriac rearrangment? (ie)
    [tex]\int \frac{{x^2}}{x^{-2}}\;dx = \int \frac{x^2}{(\frac{1}{x^2})}\;dx = \int \frac{{x^2}{x^2}}{1}\;dx = \int x^{2+2}\;dx = \int x^4\;dx[/tex]
  6. Jul 28, 2012 #5


    User Avatar
    Science Advisor

    You can do this, but the thing to be careful about is how to do substitutions, how to treat your limits and differentials, and when to know if it's ok to move things in and out of the integral itself.

    Some key tips for the above is whether something is a constant, or whether variables are independent of one another (say like if you are integrating xy^2 with respect to dx and y is independent, then you can move this outside of the integral but if it's not you can't) and also how the differentials work (i.e. the dx like terms) and how you treat these.

    The rules for doing these kind of things are usually taught in an first year university course on calculus.
  7. Jul 28, 2012 #6


    User Avatar
    Science Advisor

    My point was that you cannot treat the "d" and "[itex]d^2[/itex]" in the derivative as if they were numbers and powers of numbers. They are not.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook