Converting integration of rectangular integral to spherical.

  • #1
984
74
Homework Statement:
\begin{eqnarray}
0\leq z\leq \sqrt{x^2+y^2}\\
0\leq y\leq x\\
0\leq x\leq 1
\end{eqnarray}

##f(x,y,z)=\frac{(x^2+y^2)^\frac{3}{2}}{x^2+y^2+z^2}##
Relevant Equations:
\begin{eqnarray}
x=\rho\cos\theta\cos\phi\\
y=\rho\sin\theta\cos\phi\\
z=\rho\sin\phi
\end{eqnarray}

Answer given in book: ##\frac{\pi}{12}##
Answer derived: ##\frac{3}{16}{\pi}##
I'm going to type out my LaTeX solution later on. But in the meantime, can anyone check my work? I know it's sloppy, disorganized, and skips far more steps than I care to count, but I'd very much appreciate it. I'm not getting the answer as given in the book. I think I failed this time because I assumed that ##\phi## was the angle from the x-axis (or any other axis that can be constructed with a line on the xy-plane) to some arbitrary point on the zx-plane. When I retried it while treating ##\phi## as the angle from the z-axis to some arbitrary point on the zx-plane, I got ##\frac{3}{4}\pi(\frac{\pi}{8}-1)##. Can anyone point out anything I might have done wrong in the photograph I posted while I type out a LaTeX version of my second attempt? Thank you.
 

Attachments

  • IMG_0851.JPG
    IMG_0851.JPG
    41.2 KB · Views: 31

Answers and Replies

  • #2
984
74
1 Convert rectangular bounds of integration into spherical bounds.
2 Convert ##f## to spherical form.
3 Integrate.

===1===

\begin{eqnarray}
x=\rho\cos\theta\sin\phi\\
y=\rho\sin\theta\sin\phi\\
z=\rho\cos\phi
\end{eqnarray}

\begin{eqnarray}
0\leq z\leq \sqrt{x^2+y^2}\\
0\leq y\leq x\\
0\leq x\leq 1
\end{eqnarray}

\begin{eqnarray}
x^2+y^2&=&(\rho\cos\theta\sin\phi)^2+(\rho\sin\theta\sin\phi)^2\\
&=&(\rho^2\cos^2\theta\sin^2\phi+\rho^2\sin^2\theta\sin^2\phi)\\
&=&\rho^2\sin^2\phi(\cos^2\theta+\sin^2\theta)\\
&=&\rho^2\sin^2\phi
\end{eqnarray}

\begin{eqnarray}
z=\sqrt{x^2+y^2}\\
\rho\cos\phi=\rho|\sin\phi|\\
\rho(\cos\phi-|\sin\phi|)=0
\end{eqnarray}

We're investigating a region in the first octant, so ##\sin\phi>0##, and so ##\phi=\frac{\pi}{4}##.

\begin{eqnarray}
z=0\\
\rho\cos\phi=0
\end{eqnarray}

Hence, ##\phi=\frac{\pi}{2}##.

\begin{eqnarray}
y=x\\
\rho\sin\theta\sin\phi=\rho\cos\theta\sin\phi\\
\rho\sin\phi(\sin\theta-\cos\theta)=0
\end{eqnarray}

We are investigating the projection of the curve onto the xy-plane, so ##\phi=\frac{\pi}{2}##, and so, ##\theta=\frac{\pi}{4}##.

\begin{eqnarray}
y=0\\
\rho\sin\theta\sin\phi=0
\end{eqnarray}

Similarly, ##\theta=0##.

It is given that ##x\leq1##, and that ##y\leq1##.
It is given also, that ##z\leq\sqrt{x^2+y^2}\leq\sqrt{1^2+1^2}\leq\sqrt{2}##.

Hence, ##x^2+y^2+z^2\leq1+1+2=4=2^2##, and so ##\rho\leq2##.

Our spherical bounds are therefore:

\begin{eqnarray}
0\leq\rho\leq2\\
\frac{\pi}{4}\leq\phi\leq\frac{\pi}{2}\\
0\leq\theta\leq\frac{\pi}{4}
\end{eqnarray}

===2===

\begin{eqnarray}
f&=&\frac{(x^2+y^2)^{\frac{3}{2}}}{x^2+y^2+z^2}\\
&=&\frac{\rho^3\sin^3\phi}{\rho^2}\\
&=&\rho\sin^3\phi
\end{eqnarray}

===3===

\begin{eqnarray*}
\int_0^{\frac{\pi}{4}}\int_{\frac{\pi}{4}}^{\frac{\pi}{2}}\int_0^2(\rho\sin^3\phi)(\rho^2\sin\phi) d\rho d\phi d\theta&=&\int_0^{\frac{\pi}{4}}\int_{\frac{\pi}{4}}^{\frac{\pi}{2}}\int_0^2\rho^3\sin^4\phi d\rho d\phi d\theta\\
&=&\frac{1}{4}\rho^4|_0^2\cdot\frac{\pi}{4}\cdot\int (1-\cos^2\phi)^2 d\phi\\
&=&\frac{1}{4}2^4\cdot\frac{\pi}{4}\cdot\int(1-2\cos^2\phi+\cos^4\phi) d\phi\\
&=&\frac{1}{4}\cdot16\cdot\frac{\pi}{4}\cdot\left[\phi-\int(1+\cos2\phi) d\phi+\frac{1}{4}\int(1+\cos2\phi)^2 d\phi\right]\\
&=&4\cdot\frac{\pi}{4}\cdot[\phi-\phi-\sin2\phi+\frac{1}{4}\int(1+2\cos2\phi+\cos^2(2\phi))^2 d\phi]\\
&=&\pi\cdot\left[-\sin2\phi+\frac{1}{4}(\phi+\sin2\phi+\frac{1}{2}\phi+\frac{1}{4}\sin4\phi)\right]\\
&=&\pi\cdot\left[-\sin2\phi+\frac{1}{4}(\phi+\sin2\phi+\frac{1}{2}\phi+\frac{1}{4}\sin4\phi)\right]\\
&=&\pi\left[\frac{3}{8}\phi-\frac{3}{4}\sin2\phi+\frac{1}{16}\sin4\phi\right]|_{\frac{\pi}{4}}^{\frac{\pi}{2}}\\
&=&\frac{3}{4}\pi(\frac{\pi}{8}-1)
\end{eqnarray*}
 
Last edited:
  • #4
Delta2
Homework Helper
Insights Author
Gold Member
4,542
1,840
One thing that confuses me is that you use ##\theta## for the azimuth angle and ##\phi## for the inclination angle, while I am used the other way around, but ok I ll try to overcome this.
 
  • #5
Delta2
Homework Helper
Insights Author
Gold Member
4,542
1,840
Section 2 looks correct
Section 3 doesnt look correct but even after I do the correction (wolfram says that $$\int \sin^4\phi d\phi=\frac{3}{8}\phi-\frac{1}{4}\sin2\phi+\frac{1}{32}\sin4\phi$$
i dont get the book answer.

SO you big mistake must be in correctly determining the spherical boundaries.
 
  • #6
984
74
Okay, gotcha. I'll work on it. I very much suspect it's got something to do with the ##\rho## upper-bound.

It's gotta rely on ##\theta## and ##\phi##, I think.
 
Last edited:
  • #7
984
74
Okay, so it turns out that this problem probably wanted me to use cylindrical coordinates, instead.

I cannot fathom how it can be done with spherical coordinates.
 

Attachments

  • IMG_0852.JPG
    IMG_0852.JPG
    41.5 KB · Views: 18
  • #8
Delta2
Homework Helper
Insights Author
Gold Member
4,542
1,840
Good that you did it in cylindrical coordinates and confirmed the book answer. (Cant check you work though cause no matter how much I tried to magnify it looks faint , I just cant make out all the symbols.)

When you find time post your work with cylindrical coordinates in Latex please.
 
  • #9
984
74
I found the bounds for ##r## by looking at the projection of the surface onto the xy-plane. (See illustration in OP).

===1: Converting to cylindrical coordinates===

\begin{eqnarray}
x=1\\
r\cos\theta=1\\
r=\sec\theta
\end{eqnarray}

\begin{eqnarray}
0\leq r\leq \sec\theta\\
0\leq\theta\leq\frac{\pi}{4}\\
0\leq z\leq r
\end{eqnarray}

\begin{eqnarray}
f&=&\frac{(x^2+y^2)^\frac{3}{2}}{x^2+y^2+z^2}\\
&=&\frac{r^3}{r^2+z^2}
\end{eqnarray}

===2: Solving the integral===

\begin{eqnarray}
z=r\tan s\\
dz=r\sec^2 s\,ds\\
z^2+r^2=r^2\sec^2s\\
s&:&r\mapsto\frac{\pi}{4}\\
&:&0\mapsto0\\
\end{eqnarray}

\begin{eqnarray}
\int_0^\frac{\pi}{4}\int_0^{\sec\theta}\int_0^\frac{\pi}{4}\left(\frac{r^3}{r^2\sec^2s}\right)r(r\sec^2s\,ds)\,dr\,d\theta&=&\int_0^\frac{\pi}{4}\int_0^{\sec\theta}\int_0^\frac{\pi}{4}r^3\,ds\,dr\,d\theta\\
&=&s|_0^\frac{\pi}{4}\cdot\int_0^\frac{\pi}{4}\int_0^{\sec\theta}r^3\,dr\,d\theta\\
&=&\frac{\pi}{4}\cdot\frac{1}{4}\int_0^\frac{\pi}{4}\sec^4\theta\,d\theta\\
&=&\frac{\pi}{16}\cdot\int_0^\frac{\pi}{4}(1+\tan^2\theta)\sec^2\theta\,d\theta\\
&=&\frac{\pi}{16}\cdot\left(\tan\theta+\frac{1}{3}\tan^3\theta\right)|_0^\frac{\pi}{4}\\
&=&\frac{\pi}{16}\cdot\left(1+\frac{1}{3}\right)\\
&=&\frac{\pi}{16}\cdot\left(\frac{4}{3}\right)\\
&=&\frac{\pi}{12}
\end{eqnarray}

And I think it could be possible to solve this in spherical coordinates, if you set the upper bound of ##\rho=\sec\theta\csc\phi##.
 
  • #10
Delta2
Homework Helper
Insights Author
Gold Member
4,542
1,840
Yes well this seems correct, initially I had my doubts about the boundaries for ##\theta## but after looking through some inequalities at wolfram (namely ##0<\sin\theta<\cos\theta##) I think they are right.

I am not sure at all about the boundaries in spherical coordinates, to tell you the truth I am not so good in converting 3D domains between cartesian/spherical/cylindrical coordinates systems , that's why in my first post I checked first section 2 and 3 of yours.
 

Related Threads on Converting integration of rectangular integral to spherical.

Replies
7
Views
1K
Replies
4
Views
2K
Replies
3
Views
4K
Replies
6
Views
2K
M
  • Last Post
Replies
14
Views
1K
  • Last Post
Replies
10
Views
4K
Top