Register to reply

Implicit Differentiation

by dekoi
Tags: differentiation, implicit
Share this thread:
dekoi
#1
Oct25-05, 01:46 PM
P: n/a
Can someone check my answer (I am trying to find the second derivative) for any mistakes?
I have looked it over many times, and I've realized that my second derivative is not correct, but I cannot figure out why. Thank you.
[tex]\sqrt{x} + \sqrt{y} = 1[/tex]
[tex]\frac{1}{2\sqrt{x}} + \frac{1}{2\sqrt{y}}(y') = 0[/tex]
[tex]y' = (\frac{-1}{2\sqrt{x}})(\frac{2\sqrt{y}}{1}) = \frac{-2\sqrt{y}}{2\sqrt{x}} = \frac{-\sqrt{y}}{\sqrt{x}}[/tex]
[tex]y'' = \frac{(\frac{-y'}{2\sqrt{y}})(\sqrt{x}) + (\sqrt{y})(\frac{1}{2\sqrt{x}})}{(\sqrt{x})^2}[/tex]
[tex]y'' = \frac{\frac{-y'\sqrt{x}}{2\sqrt{y}} + \frac{\sqrt{y}}{2\sqrt{x}}}{x}[/tex]
[tex]y'' = \frac{-y'\sqrt{x}\sqrt{x} + \sqrt{y}\sqrt{y}}{2x\sqrt{x}\sqrt{y}}[/tex]
[tex]y'' = \frac{-(\frac{-\sqrt{y}}{\sqrt{x}})\sqrt{x}\sqrt{x} + \sqrt{y}\sqrt{y}}{2x\sqrt{x}\sqrt{y}}[/tex]
[tex]y'' = \frac{\sqrt{y}\sqrt{x} + \sqrt{y}\sqrt{y}}{2x\sqrt{x}\sqrt{y}}[/tex]
[tex]y'' = (\frac{\sqrt{x}\sqrt{y}+y}{2x\sqrt{x}\sqrt{y}})(\frac{\sqrt{x}\sqrt{y}} {\sqrt{x}\sqrt{y}})[/tex]
[tex]y'' = \frac{xy + y\sqrt{x}\sqrt{y}}{2x^2y}[/tex]
[tex]y'' = \frac{x + \sqrt{x}\sqrt{y}}{2x^2}[/tex]
Phys.Org News Partner Science news on Phys.org
Experts defend operational earthquake forecasting, counter critiques
EU urged to convert TV frequencies to mobile broadband
Sierra Nevada freshwater runoff could drop 26 percent by 2100
TD
#2
Oct25-05, 01:50 PM
HW Helper
P: 1,021
Your first derivative is correct, but I don't see what you mean with 0y' then... You also appear to have a lot of equal-signs, it seems a bit weird...

So we have:
[tex]\left( {\sqrt x + \sqrt y } \right)^\prime = \frac{1}{{2\sqrt x }} + \frac{{y'}}{{2\sqrt y }}[/tex]

Now, the second derivative is doing just the same, just a bit longer:
[tex]\left( {\frac{1}{{2\sqrt x }} + \frac{{y'}}{{2\sqrt y }}} \right)^\prime = \left( {\frac{1}{{2\sqrt x }}} \right)^\prime + \left( {\frac{{y'}}{{2\sqrt y }}} \right)^\prime = \cdots [/tex]
dekoi
#3
Oct25-05, 01:57 PM
P: n/a
I don't understand your first function. Is that supposed to be the second derivative?

dekoi
#4
Oct25-05, 02:00 PM
P: n/a
Implicit Differentiation

And I have a lot of equal signs because I'm trying to simplify my second derivative. I think that's where I made the mistake... in simplifying.
TD
#5
Oct25-05, 02:01 PM
HW Helper
P: 1,021
My first line is the first derivative, the second line is the beginning of the second derivative. Since it's linear, I already split it for you. The first term will be easy since it only depends on x, the second will be a bit longer.
dekoi
#6
Oct25-05, 02:04 PM
P: n/a
Well that's what I did, starting in my fourth line. I used the quotient rule on my simplified form of the first derivative, to find the second derivative. Shouldn't both methods result in the same answer?
TD
#7
Oct25-05, 02:17 PM
HW Helper
P: 1,021
Ok, I wasn't aware of the fact it was an equation at first.
It's much clearer now you've adjusted your initial post.

Line 6 seems to be correct, then:

[tex]y'' = \frac{{ - y'\sqrt x \sqrt x + \sqrt y \sqrt y }}{{2x\sqrt x \sqrt y }} = \frac{{y - y'x}}{{2x^{3/2} \sqrt y }}[/tex]
dekoi
#8
Oct25-05, 02:22 PM
P: n/a
How about after line 6?
Is the whole thing correct? Right up to the last line?
TD
#9
Oct25-05, 02:26 PM
HW Helper
P: 1,021
I took over after line 6 but I didn't realise you substituted y' again.
Anyway, it seems to correct (at least your solution, I didn't carefully check every step). It's possible to 'simplify' it one last bit, although it doesn't really matter a lot anymore:

[tex]y'' = \frac{{x + \sqrt x \sqrt y }}{{2x^2 }} = \frac{{\sqrt x + \sqrt y }}{{2x^{3/2} }}[/tex]
dekoi
#10
Oct25-05, 02:30 PM
P: n/a
This makes sense, however...

When I substitute a value of x, y, & y' into my equation on line 4 (the first y''), I get a different value for y'' than when I substitute the exact same values into my last equation for y''.

Why is that?
TD
#11
Oct25-05, 02:35 PM
HW Helper
P: 1,021
You realise that in line 4, the first derivative y' is still present in the expression for y''?
dekoi
#12
Oct25-05, 02:35 PM
P: n/a
nevermind..
dekoi
#13
Oct25-05, 02:36 PM
P: n/a
Yes I know.
dekoi
#14
Oct25-05, 02:38 PM
P: n/a
I used my values as:
x = 4
y = 1
y' = 1/2
TD
#15
Oct25-05, 02:42 PM
HW Helper
P: 1,021
Huh? Your initial equation was [itex]\sqrt x + \sqrt y = 1[/itex] but [itex]\sqrt 4 + \sqrt 1 \ne 1[/tex]...
dekoi
#16
Oct25-05, 02:51 PM
P: n/a
You are right.

I know why now...
I rearranged that formula to put it in the graphing calculator.
But I forgot to consider the fact that squares can be positive or negative. ................ Well at least I think that's my mistake.
Do you know what the graph of this function looks like?
Is it a circle? Or not?
dekoi
#17
Oct25-05, 02:56 PM
P: n/a
Never mind that......

Substitute x = 4 into the original equation and solve for y.

You will get the value "1" .
That's what I just got.

This is odd.....


Register to reply

Related Discussions
Implicit differentiation Calculus & Beyond Homework 3
Implicit Differentiation Calculus & Beyond Homework 3
Implicit differentiation Calculus & Beyond Homework 3
Implicit Differentiation Introductory Physics Homework 1
Implicit differentiation Calculus 6