Register to reply

Computing curvatures

by rsq_a
Tags: computing, curvatures
Share this thread:
May3-12, 12:36 PM
P: 110
I'm having a really hard time working with non-standard expressions for the curvature. This deals with two expressions for the curvature, one for a 2D plane curve, and the other for an axi-symmetric surface in 3D.

The plane curve

Suppose we have a plane curve given by [itex]x = h(y)[/itex] for both x and y positive. In a paper I'm reading, the author writes:
The surface curvature can be expressed in terms of the slope angle, [itex]\beta[/itex] of the surface as per
\kappa = -\frac{d( \cos\beta)}{dy}.

The geometrical relation,
\frac{dh}{dy} = \frac{1}{\tan \beta}

expresses the dependence of h on [itex]\beta[/itex]
The second expression is more or less clear for me (modulo whether it should be negated or not). The first expression is not. How do you go from the standard definition:
\kappa = \frac{h''}{(1+(h')^2)^{3/2}},

to this result?

Surface curvature

This one is from another paper. The author assumes that there is an axi-symmetric surface [itex]S(z,r) = 0[/itex], which is only a function of [itex]z[/itex] and [itex]r[/itex] in spherical coordinates. He states that the mean curvature is
\kappa = (S_z^2 +S_r^2)^{-3/2} \left[ S_z^2 S_{rr} - 2S_z S_r S_{rz} + S_r^2 S_{zz} + r^{-1} S_r(S_r^2 + S_z^2)\right].

If we denote the downward angle of the slope at an arbitrary position on the drop surface by [tex]\delta[/tex], so that
\cos\delta = \frac{S_z}{\sqrt{S_z^2+S_r^2}} \qquad
\sin\delta = \frac{S_r}{\sqrt{S_z^2+S_r^2}}

then the curvature can be written as
\kappa = \frac{1}{r} \frac{d}{dr} \left(r \sin\delta\right)

along [itex]S = 0[/itex]

Again, I could really use some help in seeing how these expressions were derived. Or if it's not trivial, perhaps a source where the work is shown.
Phys.Org News Partner Mathematics news on
Heat distributions help researchers to understand curved space
Professor quantifies how 'one thing leads to another'
Team announces construction of a formal computer-verified proof of the Kepler conjecture

Register to reply

Related Discussions
A relatively easy differential geometry question concerning principle curvatures Calculus & Beyond Homework 4
'Centripetal Accelerations' and 'Radius of Curvatures' in Projectile Motion Introductory Physics Homework 5
Lipschitz-Killing Curvature at a point on a surface in R^3 simply mean Differential Geometry 0
Plotting space-time curvatures as curved 2D-surfaces Special & General Relativity 9
Curvatures in spacetime explained by momentum? Special & General Relativity 11