How do I find the second derivative of a higher order function?

  • Thread starter Thread starter loadsy
  • Start date Start date
  • Tags Tags
    Derivative
Click For Summary

Homework Help Overview

The discussion revolves around finding the second derivative of a function defined implicitly by the equation x^4 + y^4 = a^4. Participants are exploring the process of implicit differentiation and the implications of treating constants during differentiation.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the steps involved in differentiating the implicit equation, questioning the treatment of the constant 'a' and its derivative. There is a focus on deriving dy/dx and subsequently y'' through implicit differentiation.

Discussion Status

Several participants have provided clarifications regarding the differentiation process, particularly the treatment of constants and the application of the chain rule. There is ongoing exploration of how to express y'' in terms of x, y, and y', with some participants suggesting substitutions for y' in the derived equations.

Contextual Notes

Participants are navigating through the implications of differentiating a constant and the resulting expressions, with some confusion regarding the correct forms of the derivatives and the application of implicit differentiation rules.

loadsy
Messages
57
Reaction score
0
Alright I decided I'd create a new topic just because the other one was getting fairly lengthy.

I'm having trouble with the following "Higher Derivative" question

It states, find y'' by implicit differentiation.

x^4 + y^4 = a^4

d/dx (x^4+y^4) = d/dx (a^4)
4x^3 + 4y^3 dy/dx = 4a^3

However, what is the next step from here, I thought perhaps cancelling out all the 4's and leaving it as:
x^3 + y^3 dy/dx = a^3
and bring the x^3 over so it's:
a^3-x^3 = y^3 dy/dx
dy/dx = a^3-x^3 / y^3

However, I'm not positive on that because I was told from a friend that 4a^3 is a constant so it can be equal to 0, thus the dy/dx would be -x^3/y^3.

Am I on the right track here or is there an easier way of solving this equation. Thanks a lot guys.
 
Physics news on Phys.org
If 'a' is a constant, then its derivative (with respect to any variable) is zero.
 
However, I'm not positive on that because I was told from a friend that 4a^3 is a constant so it can be equal to 0, thus the dy/dx would be -x^3/y^3.

4a^3 can't just be written as 0. But as neutrino pointed out, the derivative of a^4 isn't 4a^3
 
Ahhh I think I'm following you, so pretty much from the first step you could automatically assume that a^4 it is equal to 0 when you take the derivative, and disregard the 4a^3 because that's not possible since it is a constant, thus after solving the problem it would be dy/dx = -x^3/y^3, but that's for y', not y''. You need to take the derivative of that now if I'm not mistaken using the rule:

f'g-fg' / g^2

And sub in:

-x^3(y^3) - y^3(-x^3)
---------------------------
(y^3)^2

And take the derivatives.
 
Last edited:
Correct. In fact, I'm sure you'll notice that once you use the chain rule, you'll get y'' in terms of x, y, and y' All you need to do is substitute -x3/y3 for y'
 
loadsy said:
Ahhh I think I'm following you, so pretty much from the first step you could automatically assume that a^4 it is equal to 0 when you take the derivative, and disregard the 4a^3 because that's not possible since it is a constant, thus after solving the problem it would be dy/dx = -x^3/y^3, but that's for y', not y''. You need to take the derivative of that now if I'm not mistaken.
You're right, and you can also take the derivative x^3 + y^3 dy/dx = 0, using the product rule for the second term.
 
Notice that when you differentiated y^4 with respect to x you did not get "4y^3", you got "4y^3 y' ". You could say that the derivative of a^4 with respect to x is 4a^3 a' and then since a is a constant, a'= 0 so (a^4)'= 4a^3(0)= 0. Of course, it's simpler to argue that a^4 is itself a constant so (a^4)'= 0 immediately.

As neutrino said, once you have x^2+ y^3 y'= 0 you can just continue using "implicit differentiation": 2x+ 3y^2 (y')^2+ y^3 y"= 0
 
Oh wow, alright thanks a lot for the advice all of you. HallsofIvy clarified that constant business for me perfectly. I understand that a whole lot better now. Except one correction in your description, it's x^3+y^3 y' = 0 believe as neutrino suggested, not x^2+ y^3 y'= 0
Hence, it would be:
3x^2 + 3y^2 (y')^2 + y^3 y'' = 0

And then for y' you'd sub in -x^3/y^3 as Office Shredder suggested, and solve for y''?

So it would be 3x^2 + 3y^2(-x^3/y^3)^2 + y^3 y'' = 0 I think anyways.
 
Last edited:
Alright here's an update on what I did for this question anyways, because I assume you use the rule: f'g-fg'/g^2
so from -x^3/y^3 to differentiate this expression I went:

-3x^2(y^3) - (-x^3)(3y^2)y' / (y^3)^2
= -3x^2(y^3) - (-x^3)(3y^2)(-x^3/y^3) / (y^3)^2
Now how do I factor this expression.
 

Similar threads

  • · Replies 25 ·
Replies
25
Views
2K
  • · Replies 24 ·
Replies
24
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
Replies
3
Views
2K
  • · Replies 16 ·
Replies
16
Views
3K
Replies
4
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
Replies
2
Views
1K
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
942