Reply
Thread Tools
Math: Havin' a ball... math
Old 03-01-2011, 07:50 AM   #1
Monte314
Core Member [412%]
Chief Scientist; Adjunct Full Professor of Computer Science; Assoc. Professor of Mathematics; various national and state Advisory Panels; author of two books, many papers; Jedi Math Dog
MBTI: INTJ
Join Date: Apr 2008
Posts: 16,505
 
For a right circular cone to completely enclose a sphere, its volume must be at least X times the volume of the sphere:

....../\............................/\
For /O\ to happen, volume /..\ >= X times volume O

What is X?

(This problem can be done without the use of calculus)

[HIDE="Artistic rendering that puts Rembrandt to shame"]

To view links or images in this forum your post count must be 2 or greater. You currently have 0 posts.


[/HIDE]
Monte314 is offline
Reply With Quote

Old 03-01-2011, 08:22 AM   #2
themuzicman
Core Member [288%]
I am INTJ.  Your argument is invalid.
Resistance is futile.
MBTI: INTJ
Join Date: Jun 2009
Posts: 11,541
 

OK, the key is calculating the radius of the sphere that touches both the point and the base of the cone. This problem can be reduced to the 2d equivalent of a triangle in a circle.

If we assume a circle with radius (R) of 2, a triangle whose base intersects the center of the circle will have a base (B) of 4, and height (H) of 2. A theoretical triangle with a B = 0 will have H = 4.

Given these two points (and a rash assumption), the radius, then, should be:

(1/2B + H)/2 (Average length of the base and height)

And the area of the circle:

A = ((1/2B + H)i/2)^2 * pi.

Blowing this out to the cone and sphere, then, given the diameter of the base of the cone as D, height of cone as H, we have:

V = 4/3 * ((1/2D + H)/2)^3 * pi.

If my rash assumption is incorrect, then we'll have to dive into trigonometry...

Which I have neither the time nor patience to do ATM, although my instincts are telling me that I am to use the Cosine of one and the Sine of the other...
themuzicman is offline
Reply With Quote
Old 03-01-2011, 10:35 AM   #3
Malle
Member [06%]
 
MBTI: INTJ
Join Date: Dec 2008
Posts: 256
 
I have a long solution which does make use of calculus
Place the sphere of radius r such that it tangents the xz-plane
x² + (y-r)² = r²
Since the problem is rotationally symmetric, study the xy-plane. Place the cone (a triangle in the xy-plane) around the sphere, such that the cone has its tip at x=0, y=h and its width at it's base, where y=0, is 2b.

For this we can write the volume of the two bodies as
V_s = (4/3)(πr³)
V_c = (1/3)(πb²h)
V_r = V_c / V_s = (1/4)(b²h/r³)

Since the font isn't very good, it should be noted that π = pi

We then study all possible such triangles. We do this by studying the tangent lines to the circle for 0 < x < r.

It's straightforward to find the slope of the tangent line that tangents in the point x0:
(dy/dx)|(x=x0) = -x0 / √(r^2-x0^2)
where (f)|(x=x0) is f evaluated for x = x0

Using the equation
x² + (y-r)² = r²
and the equation for a straight line
y = (dy/dx)|(x=x0) * x + h
it's possible to solve for h which gives
h = r + r²/√(r²-x0²)

Finding where the tangent line intersects the x-axis (or xz-plane) we find our b as
b = (r² + r√(r²-x0²))/x0

Proceed to solve the equation
(d/dx0) ((1/4)(b²h/r³)) = 0
which I had to use Wolfram Alpha for, or I'd make plenty of errors and take a long time to solve it, we can find the simple solution that
x0 = r(2√2)/3

Substituting this into the equation for V_r all dependencies on r are eliminated and we end up with
V_r = 2

Thus, a cone that circumscribes a sphere must have a volume that is at least twice as large as the sphere's volume.


I'll keep thinking on a solution that does not require calculus, because this solution is not pretty.
Malle is offline
Reply With Quote
Old 03-01-2011, 11:44 AM   #4
themuzicman
Core Member [288%]
I am INTJ.  Your argument is invalid.
Resistance is futile.
MBTI: INTJ
Join Date: Jun 2009
Posts: 11,541
 
D'oh...

From the 2d version (Flatten out the cone and sphere into triangle and circle), we can form a triangle using the difference between the radius (r) and the height (cone=h) of the triangle as height (ht) and half the base (cone=b) of the triangle (bt).

(a picture would be great here, but alas... if someone can send me an email, I can send it to you so it can be uploaded to an image host and i could refer to that address... can't upload images here at work)

Thus:

ht = |h-r| (doesn't matter at this point whether the height is greater or less than, I think)
bt = 1/2 b

The hypotenuse is the radius of the circle.

Thus (using an old friend):

r^2 = bt^2 + ht^2

or

r^2 = (1/2 b)^2 + (h - r)^2

r^2 = 1/4 *b^2 + h^2 -2h * r + r^2 (uh oh!)

0 = 1/4 b^2 + h^2 - 2hr

2hr = 1/4 b^2 + h^2

r = b^2/8h + h^2/2h (combining 1/4 with 1/2h while dividing by 2h)

r = b^2/8h + h/2

Thus, the area of the sphere is:

V= 4/3 (b^2/8h + h/2) * pi^3

(maybe?)
themuzicman is offline
Reply With Quote
Old 03-01-2011, 12:56 PM   #5
BingeArtist
Member [08%]
MBTI: INTJ
Join Date: Apr 2010
Posts: 345
 
No calculus.


Answer: X = 2.

Solution: We assume the sphere is radius 1. Let h and r be the height and radius of the cone. What we want to do is minimize r^2*h.

In the plain (first quadrant..I think..with both x & y nonnegative), we can represent the sphere by the equation x^2 + (y-1)^2 = 1, and we can represent the cone by the line y = -(h/r)x + h.

Let P be the point where the line (tangently) intersects the circle. From similar triangles, we have D(P, (0,h)) = h/r. Applying the Pythagorean Theorem to the right triangle defined by P, (0,h), and (0,1), we have (h/r)^2 + 1 = (h-1)^2. Hence, r^2 = h/(h-2). So, since we're trying to minimize r^2*h, we're trying to minimize h^2/(h-2).

Here's where we have to use a few tricks to avoid the dreaded calculus. To minimize h^2/(h-2), we want to maximize (h-2)/h^2 = 1/h - 2/(h^2). Set m = 1/h. Hence, we're trying to maximize m - 2m^2. And this is just a parabola with a vertex at (1/4, 1/8). So our minimum value of r^2*h is 8, and occurs when h = 4 (and hence, when r^2 = 2).

So the volume of the cone is 8 pi over 3, and the volume of the sphere is 4 pi over 3. So the ratio is 2:1. ###
BingeArtist is offline
Reply With Quote
Old 03-02-2011, 11:33 AM   #6
K27
Member [28%]
Canon EOS 600D 1.8 50mm
MBTI: INTJ
Join Date: Apr 2009
Posts: 1,151
 
[HIDE="spoiler"]Let the radius of the base of the cone be b, height of the cone be h and radius of the sphere be b.
Looking at the intersection of vertical half of the cone and the sphere, and drawing a line from the centre of the sphere to the touching point of the sphere to the cone, we obtain two similar triangle.

By similar triangle, we get:
r / (h - r) = b / sqr(h^2 - b^2)

Expend and do some arithmetic, we get:
b^2 = (r^2)h / (h - 2r) ..... (1)

Substitute into the volume of cone, we get:
(1/3)(pi)(r^2)[h^2/(h-2r)]

We have to minimise h^2/(h-2r). I initially used calculus, but after peeking into BingeArtist's solution I did the same thing... Maximise (h - 2r) / h^2 and take x = 1/h... Then got h = 4r in my solution.

Then sub h = 4r into (1) we get b^2 = 2r^2.

So the volume of the cone is:
(1/3)(pi)(b^2)h
= (1/3)(pi)(2r^2)(4r)
= (8/3)(pi)(r^3)
= 2(4/3)(pi)(r^3)

Therefore, X=2![/HIDE]

I think BingeArtist's way of avoiding calculus is brilliant!
K27 is offline
Reply With Quote
Old 03-02-2011, 11:49 AM   #7
themuzicman
Core Member [288%]
I am INTJ.  Your argument is invalid.
Resistance is futile.
MBTI: INTJ
Join Date: Jun 2009
Posts: 11,541
 
to finish:

V(s) = 4/3 (b^2/8h + h/2) * pi^3 volume of Sphere
V(c) = 1/3 pi * b^2 * h

So, the ratio (X) is:

X=V(s)/V(c)

or

X= [4/3 (b^2/8h + h/2)^3 * pi] / [1/3 pi * b^2 * h]

(reducing a bit)

X=[4(b^2/8h + h/2)^3 ] / [b^2 * h]


themuzicman is offline
Reply With Quote
Old 03-02-2011, 01:59 PM   #8
envirodude
Core Member [112%]
What makes the desert beautiful, said the little prince, is that somewhere it hides a well.
MBTI: INTx
Join Date: Dec 2010
Posts: 4,517
 
I got the answer, 2, by drawing a section through the centre of the cone&sphere, and defining alpha as the base angle of the isoceles triangle formed. Simple trig and similar triangles leads to the equation for the volume ratio as:

X = (cos alpha +1)³ / ( sin² alpha · cos alpha)

a bit of thinking leads to the constraint: 0 < alpha < 90°

To minimise X, I used Solver in Excel, and confirmed with Wolfram Alpha [minimize (cos a +1)^3/(sin a)^2/cos a]. That's not calculus, is it?

alpha = 70.5287°; X = 2
envirodude is offline
Reply With Quote
Old 03-02-2011, 02:18 PM   #9
Monte314
Core Member [412%]
Chief Scientist; Adjunct Full Professor of Computer Science; Assoc. Professor of Mathematics; various national and state Advisory Panels; author of two books, many papers; Jedi Math Dog
MBTI: INTJ
Join Date: Apr 2008
Posts: 16,505
 
Nice use of automated tools!
Monte314 is offline
Reply With Quote
Old 03-02-2011, 08:24 PM   #10
rufsketch1
Member [27%]
MBTI: INTP
Join Date: Dec 2008
Posts: 1,102
 
Oh that's easy. The ratio of the cone's volume to that of the sphere is V/(4r³/pi) : 1

Where V is the lowest point between 0 and 1.57 on the function

To view links or images in this forum your post count must be 2 or greater. You currently have 0 posts.


---------- Post added 03-02-2011 at 11:32 PM ----------

Or as you earthlings would say: "2".
rufsketch1 is offline
Reply With Quote
Old 03-02-2011, 09:50 PM   #11
Monte314
Core Member [412%]
Chief Scientist; Adjunct Full Professor of Computer Science; Assoc. Professor of Mathematics; various national and state Advisory Panels; author of two books, many papers; Jedi Math Dog
MBTI: INTJ
Join Date: Apr 2008
Posts: 16,505
 
Wow; it's obvious when you put it that way!

*Dog begins hyperventilating*
Monte314 is offline
Reply With Quote
Reply

Tags
math

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 09:12 PM.


Powered by vBulletin®
Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.
Myers-Briggs Type Indicator, Myers-Briggs, and MBTI are trademarks or registered trademarks of the
Myers-Briggs Type Indicator Trust in the United States and other countries.