# Plane intersecting a sphere

The unit sphere x2 + y2 + z2 =1 with density =1 is cut by the plane z=1/2. Find the volume and centroid of each piece.

As of now, i have (for the top piece) the integral of $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt, with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 1/(2Cos[phi]) to 1, but when i multiply in rCos[p] to find the z coordinate of the centroid, I get 3/8 which is below that part of the sphere.

And as far as the lower piece of the sphere goes, I can't figure out what the limits on my triple integrals would be. I can find the volume by seperating it in to two pieces like this:

From the xy plane up to z=1/2: $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 0 to 1/(2Cos[phi]).

From xy plane down (the lower hemisphere): $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 0 to 1.

Add these together gives the volume, but what would I do to find the centroid of the entire piece?

By the way, r= row, p= phi, and t= theta

## Answers and Replies

HallsofIvy
Science Advisor
Homework Helper
The unit sphere x2 + y2 + z2 =1 with density =1 is cut by the plane z=1/2. Find the volume and centroid of each piece.

As of now, i have (for the top piece) the integral of $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt, with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 1/(2Cos[phi]) to 1, but when i multiply in rCos[p] to find the z coordinate of the centroid, I get 3/8 which is below that part of the sphere.
$\phi$ does NOT go from 0 to $\pi/2$. The plane z= 1/2 cuts the unit sphere where $x^2+ y^2+ 1/4= 1$ or $x^2+ y^2= 3/4$. The distance from (0,0,0) to, say, $(\sqrt{3}{2}, 0, 1/2)$ on that circle is 1 so we have $\rho= 1$ on that circle. Since $x^2+ y^2= 3/4$ in spherical coordinates is $\rho^2 sin^2(\phi)= 3/4$, with $\rho= 1$, we have $sin^2(\phi)= 3/4$ so $sin(\phi)= \sqrt{3}/2$. $\phi$ goes from 0 to $\pi/3[itex]. And as far as the lower piece of the sphere goes, I can't figure out what the limits on my triple integrals would be. I can find the volume by seperating it in to two pieces like this: From the xy plane up to z=1/2: $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 0 to 1/(2Cos[phi]). From xy plane down (the lower hemisphere): $$\int$$$$\int$$$$\int$$ r2 Sin[p] dr dp dt with theta from 0 to 2 pi, phi from 0 to pi/2, and row from 0 to 1. Add these together gives the volume, but what would I do to find the centroid of the entire piece? By the way, r= row, p= phi, and t= theta (The standard english transliteration of the Greek letter [itex]\rho$ is "rho".)

Actually, a better way to do it is to integrate on the cone, $\phi$ from 0 to $\pi/3$, $\rho$ from 0 to $1/(2cos(\phi))$ and $\theta$ from 0 to $2\pi$, then the rest of the sphere: $\phi$ from $\pi/3$ to $\pi$, $\rho$ from 0 to 1, and $\theta$ from 0 to $2\pi$.

As for the centroid do exactly the same thing: multiply dV by $z= \rho cos(\phi)$, integrate over those two ranges and add. Then divide by the volume.