D/dx in Spherical Coordinates: What am I Missing?

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
6 replies · 13K views
raul_l
Messages
105
Reaction score
0

Homework Statement



Hi. I have a simple question. Is it true that [tex]\frac{\partial r}{\partial x} = (\frac{\partial x}{\partial r})^{-1}[/tex] ?

Because I'm having some trouble with the conversion between rectangular and spherical coordinates.

Homework Equations



[tex]x = r cos \phi sin \theta[/tex]

[tex]y = r sin \phi sin \theta[/tex]

[tex]z = r cos \theta[/tex]

[tex]r = \sqrt{x^2+y^2+z^2}[/tex]

The Attempt at a Solution



It is easy to show that
[tex]\frac{\partial r}{\partial x} = cos \phi sin \theta[/tex]

However, we see that
[tex](\frac{\partial x}{\partial r})^{-1} = (\frac{\partial (r cos \phi sin \theta)}{\partial r})^{-1}= \frac{1}{cos \phi sin \theta}[/tex]

and these are clearly not equal.

What am I missing? :confused:
 
Physics news on Phys.org
Hi raul_l! :smile:

(have a curly d: ∂ :wink:)
raul_l said:
Hi. I have a simple question. Is it true that [tex]\frac{\partial r}{\partial x} = (\frac{\partial x}{\partial r})^{-1}[/tex] ?

No, it doesn't work for partial derivatives, because they depend on what the other (unwritten) coordinates are.

∂r/dx keeps y constant, but ∂x/dr keeps θ constant …

and keeping y and θ constant aren't the same! :wink:
 
Yes, I suspected that. Thanks.

But it's weird that if I multiply both sides by [tex]\frac{\partial x}{\partial r}[/tex] I get
[tex]\frac{\partial r}{\partial x} \frac{\partial x}{\partial r} = 1[/tex]
and that seems mathematically correct.
 
You made a little mistake.:wink:

[tex]\frac{dx}{dr}=cos \phi sin \theta[/tex]

Proof.

[tex] r = \frac{x}{cos \phi sin \theta} [/tex]

[tex]dr=\frac{(x)'cos \phi sin \theta - x(cos \phi sin \theta)'}{cos^2 \phi sin^2 \theta} dx[/tex]

[tex]dr=\frac{cos \phi sin \theta}{cos^2 \phi sin^2 \theta}dx[/tex]

[tex]dr=\frac{1}{cos \phi sin \theta}dx[/tex]

[tex]\frac{dr}{dx}=\frac{1}{cos \phi sin \theta}[/tex]

Is this right, mate :smile: ?
 
Last edited:
njama said:
Is this right, mate :smile: ?

Sorry, chuck :redface:

not unless you're using those weird (x,θ,φ) coordinates. :rolleyes:
 
Just you need to prove that

[tex]\frac{dr}{dx} = (\frac{dx}{dr})^{-1}[/tex]

[tex](\frac{dx}{dr})^{-1}=(cos \phi sin \theta)^{-1}=\frac{1}{cos \phi sin \theta}=\frac{dr}{dx}[/tex]

I can do science me. :smile:
 
njama, there's a difference between [tex]\frac{dr}{dx}[/tex] and [tex]\frac{\partial r}{\partial x}[/tex].

Correct me if I'm wrong but I think the left side of what you wrote is equal to
[tex]\frac{dr}{dx} = \frac{\partial r}{\partial x} + \frac{\partial r}{\partial y} \frac{d y}{d x} + \frac{\partial r}{\partial z} \frac{d z}{d x}[/tex]

while right side is equal to
[tex](\frac{dx}{dr})^{-1} = (\frac{\partial x}{\partial r} + \frac{\partial x}{\partial \phi} \frac{d \phi}{d r} + \frac{\partial x}{\partial \theta} \frac{d \theta}{d r})^{-1}[/tex]

and it doesn't look like these would be equal.
 
Last edited: