Ellipse's Circular Tangents

Geometry Level 5

Inside an ellipse, two circles are drawn which touch the minor axis of the ellipse and also touch the ellipse at the ends of latus rectum. The combined area of the two circles is A 1 A_1 and area of ellipse is A 2 A_2 .

If A 1 A 2 = a b + c where a , b , c N and gcd ( a , b , c ) = 1 \displaystyle \frac{A_1}{A_2} = \frac{a}{\sqrt{b} + c} \text{ where } a, b, c \in \N \text{ and } \gcd(a,b,c) = 1 , enter answer as a 2 + b 2 + c 2 a^2 + b^2 + c^2 .


Try similar problems

All of my problems are original


The answer is 50.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Aryan Sanghi
Sep 25, 2020

Let's solve it using coordinate geometry

Let the equation of ellipse be x 2 a 2 + y 2 b 2 = 1 \displaystyle \frac{x^2}{a^2} + \frac{y^2}{b^2} = 1 where a > b a \gt b . Let eccentricity of ellipse be e e .

Now, let the parametric coordinates of points on ellipse be ( a cos θ , b sin θ ) (a \cos \theta, b \sin\theta) .

Now, coordinates of endpoints of latus rectum are ( a e , b 2 a ) \displaystyle \bigg(ae, \frac{b^2}{a}\bigg) and ( a e , b 2 a ) \displaystyle \bigg(ae, -\frac{b^2}{a}\bigg) . Comparing with polar coordinates we get cos θ = e and sin θ = ± 1 e 2 \cos \theta = e \text{ and } \sin\theta = \pm\sqrt{1 - e^2}

Now, equation of normal at point ( a cos θ , b sin θ ) (a \cos \theta, b \sin\theta) is a x cos θ b y sin θ = a 2 b 2 \displaystyle \frac{ax}{\cos \theta} - \frac{by}{\sin \theta} = a^2 - b^2

So, equations of normals at endpoints of latus rectum are

a x e b y 1 e 2 = a 2 b 2 and a x e + b y 1 e 2 = a 2 b 2 \frac{ax}{e} - \frac{by}{\sqrt{1-e^2}} = a^2 - b^2 \text{ and } \frac{ax}{e} + \frac{by}{\sqrt{1-e^2}} = a^2 - b^2

Now the normals will intersect at circle's center C ( h , k ) C(h, k) . Solving, we get

h = e a 2 b 2 a and k = 0 h = e\frac{a^2 - b^2}{a} \text{ and } k = 0

Now, b 2 = a 2 ( 1 e 2 ) b^2 = a^2(1-e^2) , so

h = a e 3 and k = 0 \boxed{h = ae^3 \text{ and } k = 0}


Now, as circle touches minor axis, i.e. y-axis, so

radius of circle = x-coordinate of center \text{radius of circle } = \text{x-coordinate of center}

radius of circle = a e 3 \boxed{\text{radius of circle } = ae^3}

Now, radius of circle = distance between center of circle and ends of latus rectum \text{radius of circle } = \text{distance between center of circle and ends of latus rectum}

a e 3 = ( a e 3 a e ) 2 + ( b 2 a ) 2 ae^3 = \sqrt{(ae^3 - ae)^2 + \bigg(\frac{b^2}{a}\bigg)^2}

a e 3 = ( a e 3 a e ) 2 + ( a ( 1 e 2 ) ) 2 ae^3 = \sqrt{(ae^3 - ae)^2 + (a(1-e^2))^2}

a 2 e 6 = a 2 e 6 2 a 2 e 4 + a 2 e 2 + a 2 2 a 2 e 2 + a 2 e 4 a^2e^6 = a^2e^6 - 2a^2e^4 + a^2e^2 + a^2 - 2a^2e^2 + a^2e^4

e 4 + e 2 1 = 0 e^4 + e^2 - 1 = 0

e 2 = 5 1 2 \boxed{e^2 = \frac{\sqrt5 - 1}{2}}

Now, area of circles A 1 = 2 × ( π ( a e 3 ) 2 ) = 2 π a 2 e 6 A_1 = 2×(\pi (ae^3)^2)= 2\pi a^2e^6 and area of ellipse A 2 = π a b = π a 2 1 e 2 A_2 = \pi ab = \pi a^2\sqrt{1-e^2}

A 1 A 2 = 2 π a 2 e 6 π a 2 1 e 2 = 2 e 6 1 e 2 = 2 ( 5 1 2 ) 3 1 5 1 2 = 2 5 2 6 2 5 4 = 4 5 2 5 1 = 4 3 + 5 \begin{aligned} \frac{A_1}{A_2} & = \frac{2\pi a^2e^6}{\pi a^2 \sqrt{1-e^2}} \\ & = \frac{2e^6}{\sqrt{1-e^2}} \\ & = \frac{2(\frac{\sqrt5 - 1}{2})^3}{\sqrt{1-\frac{\sqrt5 - 1}{2}}} \\ & = 2\frac{\sqrt5 - 2}{\sqrt{\frac{6-2\sqrt5}{4}}} \\ & = 4\frac{\sqrt5 - 2}{\sqrt5 - 1} \\ & = \boxed{\frac{4}{3 + \sqrt5}} \end{aligned}


So, a = 4 , b = 5 , c = 3 , a 2 + b 2 + c 2 = 50 a = 4, b = 5, c = 3, a^2 + b^2 + c^2 = 50

Visualisation: Area of circles is about 76.393 % 76.393\% of ellipse's area

@Aryan Sanghi a very nice problem and solution. Thanks for posting. Upvoted

Talulah Riley - 8 months, 2 weeks ago

Log in to reply

Thanku for appreciation. :)

Aryan Sanghi - 8 months, 2 weeks ago

Log in to reply

@Aryan Sanghi You deserve appreciation bro :)

Talulah Riley - 8 months, 2 weeks ago

@Aryan Sanghi , you can align the equal sign = = using \begin{align} \end{align}. Note that double backslash \ \ \backslash \backslash is to go to next line.

Chew-Seong Cheong - 8 months, 2 weeks ago

Log in to reply

@Chew-Seong Cheong I didn't understand, the application?

Talulah Riley - 8 months, 2 weeks ago

Log in to reply

They are LaTex codes. I have used them to align the equal sign = = 's after A 1 A 2 \dfrac {A_1}{A_2} in Aryan's solution.

Chew-Seong Cheong - 8 months, 2 weeks ago

Ohk, I didn't knew it. I'll definitely use it from next solution. :)

Aryan Sanghi - 8 months, 2 weeks ago

Log in to reply

You can size your image by adding after jpg "|width=400". The large the number, for example 500, the larger the image. Check the image in your solution. For the problem, I added width=300.

Chew-Seong Cheong - 8 months, 2 weeks ago

You must align with &

Chew-Seong Cheong - 8 months, 2 weeks ago
Chew-Seong Cheong
Sep 26, 2020

Interestingly, the ellipse that satisfies the conditions is x 2 φ 2 + y 2 = 1 \dfrac {x^2}{\varphi^2} + y^2 = 1 , where the major axis a = φ = 1 + 5 2 a = \varphi = \dfrac {1+\sqrt 5}2 is the golden ratio and the minor axis b = 1 b = 1 or an eccentricity e = 1 φ e = \dfrac 1{\sqrt \varphi} (see proof below). Then the radius of the two circle is r = e = 1 φ r = e = \dfrac 1{\sqrt \varphi} . Then A 1 A 2 = 2 π r 2 π a b = 2 φ 2 = 2 ( 1 + 5 2 ) 2 = 8 2 5 + 6 = 4 5 + 3 \dfrac {A_1}{A_2} = \dfrac {2\pi r^2}{\pi a b} = \dfrac 2{\varphi^2} = \dfrac 2{\left(\frac {1+\sqrt 5}2\right)^2} = \dfrac 8{2\sqrt 5+6} = \dfrac 4{\sqrt 5 + 3} . Therefore a 2 + b 2 + c 2 = 4 2 + 5 2 + 3 2 = 50 \bold a^2 + \bold b^2 + \bold c^2 = 4^2 + 5^2 + 3^2 = \boxed{50} . (Note that a \bold a and b \bold b are not major and minor axes.)

Proof: Let the ellipse be x 2 a 2 + y 2 b 2 = 1 \dfrac {x^2}{a^2} + \dfrac {y^2}{b^2} = 1 with foci F 1 ( c , 0 ) F_1(-c,0) and F 2 ( c , 0 ) F_2(c,0) , where c = a 2 b 2 c = \sqrt{a^2-b^2} .

Note that latus rectum is the straight line passes through a focus and perpendicular ot the major axis. Let the top end of the latus rectum through F 2 F_2 be P ( c , v ) P(c, v) . By definition of ellipse F 1 P + F 2 P = 2 a F_1P + F_2P = 2a or

4 c 2 + v 2 + v = 2 a 4 c 2 + v 2 = 2 a v 4 c 2 + v 2 = 4 a 2 4 a v + v 2 v = a 2 c 2 a = b 2 a \begin{aligned} \sqrt{4c^2+v^2} + v & = 2a \\ \sqrt{4c^2+v^2} & = 2a - v \\ 4c^2 + v^2 & = 4a^2 - 4av + v^2 \\ \implies v & = \frac {a^2-c^2}a = \frac {b^2}a \end{aligned}

The gradient at a point on the ellipse is given by d y d x = b 2 x a 2 y \dfrac {dy}{dx} = - \dfrac {b^2x}{a^2y} . At P ( c , v ) P(c,v) , we have d y d x x = c = b 2 c a 2 v = c a \dfrac {dy}{dx}\bigg|_{x=c} = - \dfrac {b^2c}{a^2v} = - \dfrac ca . Let the radius of the circles be r r . Then the circle for x 0 x \ge 0 is ( x r ) 2 + y 2 = r 2 (x-r)^2 + y^2 = r^2 and d y d x = x r y \dfrac {dy}{dx} = - \dfrac {x-r}y . At P ( c , v ) P(c,v) , d y d x x = c = c r v \dfrac {dy}{dx}\bigg|_{x=c} = - \dfrac {c-r}v . Since the circle and ellipse are tangent at P P , then

c r v = c a a ( c r ) b 2 = c a c r = b 2 a 2 c r = c ( 1 b 2 a 2 ) = c 3 a 2 \begin{aligned} \frac {c-r}v & = \frac ca \\ \frac {a(c-r)}{b^2} & = \frac ca \\ c - r & = \frac {b^2}{a^2}c \\ r & = c \left(1- \frac {b^2}{a^2} \right) = \frac {c^3}{a^2} \end{aligned}

For the circle at P P :

( c r ) 2 + v 2 = r 2 c 2 2 c r + r 2 + b 4 a 2 = r 2 c 2 2 c 4 a 2 + b 4 a 2 = 0 a 2 c 2 2 c 4 + b 4 = 0 a 2 ( a 2 b 2 ) 2 ( a 4 2 a 2 b 2 + b 4 ) + b 4 = 0 a 4 3 a 2 b 2 + b 4 = 0 \begin{aligned} (c-r)^2 + v^2 & = r^2 \\ c^2 - 2cr + r^2 + \frac {b^4}{a^2} & = r^2 \\ c^2 - \frac {2c^4}{a^2} + \frac {b^4}{a^2} & = 0 \\ a^2c^2 - 2c^4 + b^4 & = 0 \\ a^2(a^2-b^2) - 2(a^4 - 2a^2b^2 + b^4) + b^4 & = 0 \\ a^4 - 3a^2b^2 + b^4 & = 0 \end{aligned}

a 2 = 3 b 2 + 9 b 4 4 b 4 2 = 3 + 5 2 b 2 Note that φ 2 = ( 1 + 5 2 ) 2 = 3 + 5 2 a = φ b \begin{aligned} a^2 & = \frac {3b^2 + \sqrt{9b^4-4b^4}}2 = \frac {3+\sqrt 5}2b^2 & \small \blue{\text{Note that }\varphi^2 = \left(\frac {1+\sqrt 5}2\right)^2 = \frac {3+\sqrt 5}2} \\ \implies a & = \varphi b \end{aligned}

Putting b = 1 b=1 , a = φ \implies a = \varphi , c = φ 2 1 = φ + 1 1 = φ c = \sqrt{\varphi^2 -1} = \sqrt{\varphi +1-1} = \sqrt \varphi , and r = c 3 a 2 = 1 φ r = \dfrac {c^3}{a^2} = \dfrac 1{\varphi} .

Excellent solution sir. Thanku for sharing it with us. :)

Aryan Sanghi - 8 months, 2 weeks ago

Log in to reply

Nice problem.

Chew-Seong Cheong - 8 months, 2 weeks ago

Log in to reply

Thanku very much sir. :)

Aryan Sanghi - 8 months, 2 weeks ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...