Let R > 1 be the ratio of diameters of successive circles in an infinite self-similar sequence.
Let n be any n -th circle in the sequence. Then the following statements are true:
Find ⌊ 1 0 0 0 0 × R ⌋ .
Bonus : Find exact answer, which is short and pretty.
Note : A self-similar geometrical sequence is congruent to itself, after scaling, translation, and rotation.
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.
You have set this as an excellent geometry question that I haven't met before and your answer is tremendously good!
I noticed interesting property. 4 consecutive circles belong to Apollonian circles family with centers of circle 2 and 3 having same distance between them as distance between centers of circles 1 and 4 . The axis of these circle centers are perpendicular. Radius of circle 3 is equal to circumradius of triangle with vertices on centers of triangles 2, 3 4. Circumcenter of this triangle is a point of intersection of circle 1 and 3.
Very interesting question.
Log in to reply
I wonder if this is related to Xuming Liang's comment, "For example, can you show that the centers of circles n , n + 1 , n + 3 , n + 4 are concyclic?", which was posted in Featured Member-Xuming Liang ?
This problem is actually a simpler spin-off of another one that involves 6 tangent circles in a self-similar sequence, which I solved some years ago. This version is easier to post in here, but it does look like now that it does have interesting properties I hadn't caught on.
Log in to reply
I am not sure if it is related, but I think points of intersections of circles n with n + 1 , and n + 3 with n + 4 are concyclic as well. The circles on which the centers of circles n , n + 1 , n + 3 , n + 4 lie and their intersections have radius ratio of 3 .
This is a five circles' question. After all, it seems to turn for about 1 5 ∘ clockwise (to common convention), for every repeated enlargement. Named A, B, C, D and E from small to big, circle C coincides with circle A and circle E with b 2 − 4 a c = 0 confirms for a correct answer as a given clue. Set a good convention to solve this question numerically with geometry particularly for circle D and circle E, with Excel which can handle difficult figures:
1 2 3 4 5 6 7 |
|
What a golden ratio related answer! Checked that 2.89005363826396 to be correct and realized.
R = 1.700015775886789767192193615058+ and ⌊ 1 0 0 0 0 R ⌋ = 17000
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
S o l v i n g r e s o u r c e s :
A (0, 0) of radius 1
B (0, - 1 + R 2 ) of radius R
C (R 1 + 1 + R 2 R 4 , 1 + R 2 − 1 ) of radius R 2
We ought to first solve and visualize for center of circles A, B and C above.
With x 2 + y 2 = 1 centered at A (0, 0), B vertically below it would have a distance of 1 2 + R 2 , because tangent of circle A to center B ought to have a length of R which is the radius of circle B which is equals to the ratio to be determined. Since it forms a right angle triangle, distance A B and hence coordinates of B are determined as (0, - 1 + R 2 ). Circle B is x 2 + ( y + 1 + R 2 ) 2 = R 2 and we ought to point out center of circle C at C by drawing arcs to intercept but mathematically as:
x 2 + y 2 = 1 2 + R 4 and
x 2 + ( y + 1 + R 2 ) 2 = R 2 + R 4
⟹ ( y + 1 + R 2 − y ) ( y + 1 + R 2 + y ) = R 2 − 1 by subtracting equation 2 with equaton1.
With working, y = 1 + R 2 − 1 and hence x = R 1 + 1 + R 2 R 4 . C (R 1 + 1 + R 2 R 4 , 1 + R 2 − 1 ) of radius R 2 is found.
So far, ± of ? are chosen according to diagram. For D and E to come, we should choose only − ? for wanted cases. To intercept for D:
( x − R 1 + 1 + R 2 R 4 ) 2 + ( y + 1 + R 2 1 ) 2 = R 4 + R 6 and
x 2 + ( y + 1 + R 2 ) 2 = R 2 + R 6
Expand both equations and then subtract the first by the second. We shall obtain y = - x R 2 + R 2 1 + 1 .
Substitute y = - x R 2 + R 2 1 + 1 into expanded 2nd equation of x 2 + y 2 + 2 1 + R 2 y + 1 = R 6 , a quadratic equation is obtained:
( R 2 + R 2 1 + 2 ) x 2 − 2 ( 1 + R 2 ) 2 + 1 + R 2 1 x + 1 − R 6 = 0 .
Let this complicated form be a x 2 + b x + c = 0 ,
Take negative, x = 2 a − b − b 2 − 4 a c and y = - x R 2 + R 2 1 + 1 . This is the numerical value of D which we can name it is D (-p, q).
Together with C (r, -s), of which p, q, r and s are taken as positive numbers to avoid ambiguity by choosing of signs as we have now known their values, arcs are drawn again to intercept for E:
( x + p ) 2 + ( y − q ) 2 = R 6 + R 8 and
( x − r ) 2 + ( y + s ) 2 = R 4 + R 8
Numerical instead of exact is becoming more demanding up to here as forms are complicated.
With y = u x + v = q + s p + r x + 2 1 q + s p 2 + q 2 − r 2 − s 2 + R 4 − R 6 ,
( 1 + u 2 ) x 2 + 2 ( u s + u v − r ) x + r 2 + v 2 + 2 s v + s 2 − R 4 − R 8 = 0 .
Again, we take negative, x = 2 a − b − b 2 − 4 a c and y = u x + v where u and v are dynamic values with respect to R being treated as constant, or a dynamic constant. In real access, value of p, q, r and s are q u o t e d f r o m cell of -p, q, r, -s for easy references.
Numerical coordinates of E in d y n a m i c form with respect to R is obtained! To look for b 2 − 4 a c = 0 for one intercepting point between circle E and circle A:
x 2 + y 2 = 1 and
( x − a ) 2 + ( y − b ) 2 = ( R 4 ) 2
We can obtain a x + b y = 2 a 2 + b 2 + 1 − R 8 = k
Rearrange for y = b k − a x ,
x 2 + ( b k − a x ) 2 = 1
⟹ ( a 2 + b 2 ) x 2 − 2 k a x + k 2 − b 2 = 0 .
Set another b 2 − 4 a c as a responsive reference for R to look for a zero!
We shall find that only one excel value of R = 1.70001577588679 can make a zero.
This solved for R .
We shall also find that ( a 2 + b 2 k a , a 2 + b 2 k b ) are coordinates of the only touch point which can also satisfy circle C; this c o n f i r m e d that R = 1.70001577588679 is a correct finding. X (0.479144314051276, 0.877736137066449) for circle of radius = 2.89005363826396 suits.
Here ends. I found from author of this question that it is actually a golden ratio related value. His solution is an analytical result to solution like my solution here. I didn't apply rules for orthogonal interception for circles.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Answer: 1 7 0 0 0
I have actually measured and found an average of 1.7 in the beginning while trying to relate to 2 5 + 1 . Thought that the diagram can only give a rough idea for not so perpendicular's interceptions. I turned for 9 0 ∘ to measure both orientations.
Problem Loading...
Note Loading...
Set Loading...
The exact answer is R = φ + φ = 1 . 7 0 0 0 1 5 7 7 5 8 8 . . . . where φ = 2 1 ( 1 + 5 ) is the golden ratio.
Edit: Here's one method of computing R
Let r be the radius of circle n , and let 1 be the distance between centers of circle n and n + 1 , and let R be the ratio we seek. Thus, if circles n and n + 1 are orthogonal, we have
r = 1 + R 2 1
Let θ be the angle between the lines from circle n to n + 1 and from circle n + 1 to n + 2 . Since circles n and n + 2 are also orthogonal, we have, using the Law of Cosines
θ = A r c C o s ( 1 + R 2 R )
From this, and the property of self-similarity, we can compute the distance between the centers of circle n and n + 4 , which is
( k ∑ 4 R k − 1 C o s ( k ( π − θ ) ) ) 2 + ( k ∑ 4 R k − 1 S i n ( k ( π − θ ) ) ) 2 = ( 1 + R 2 1 + R 4 ) 3
Since the radius of circle n + 4 is R 4 r , and if circles n and n + 4 are tangent, then
1 + R 2 1 + ( 1 + R 2 1 + R 4 ) 3 = 1 + R 2 R 4
for which the only real positive solution is φ + φ
The fact that the point where circles n and n + 4 are tangent happens to be the point where circles n and n + 2 intersect is a bonus and not necessary to find solution.