My Math Forum Area 1 circles on radius 1 sphere

 Geometry Geometry Math Forum

 June 23rd, 2018, 01:16 PM #1 Senior Member   Joined: May 2015 From: Arlington, VA Posts: 444 Thanks: 29 Math Focus: Number theory Area 1 circles on radius 1 sphere What is the maximum number of circles of area one packable, without overlapping, onto a sphere of radius one? (I couldn't figure out StackExchange's use of tags.)
 June 23rd, 2018, 01:44 PM #2 Global Moderator   Joined: May 2007 Posts: 6,807 Thanks: 717 Do you mean on the surface of the sphere? For stack exchange "geometry" would be a tag. Thanks from Loren
June 23rd, 2018, 03:44 PM   #3
Senior Member

Joined: May 2015
From: Arlington, VA

Posts: 444
Thanks: 29

Math Focus: Number theory
Quote:
 Originally Posted by mathman Do you mean on the surface of the sphere? For stack exchange "geometry" would be a tag.
Yes, thank you, the surface. Tag, you're it!

 June 24th, 2018, 06:41 AM #4 Global Moderator   Joined: Dec 2006 Posts: 20,937 Thanks: 2210 Interesting question! Is each circle's area the area of the smaller part of the sphere's surface that is enclosed by the circle? If so, the answer must be less than 4pi (the surface area of the sphere), and so is at most 12. I would imagine that 7 non-overlapping circles could be accommodated quite easily. I think the answer is 10. Thanks from Loren and SDK
 June 24th, 2018, 07:30 AM #5 Senior Member   Joined: Sep 2016 From: USA Posts: 635 Thanks: 401 Math Focus: Dynamical systems, analytic function theory, numerics The question is very ambiguous. What does it mean to have a circle on the boundary of a sphere? To start with, you probably don't mean circles since they are 1 dimensional and have no area. I guess you mean disc? But this also has problems since strictly speaking, no disc of positive radius exists on the surface of a sphere. The only way I can parse this is that you mean a topological disc (i.e. a set which is homeomorphic to a disc) of area 1. In this case however, the answer is pretty trivial. Either you allow them to intersect in which case the answer is clearly infinite. Or, they may not intersect and skipjack's upper bound of 12 is obviously attainable. Since I doubt the latter interpretation is what you have in mind, but I don't know how else to even interpret this, you should ask the question more clearly. Thanks from Loren
June 24th, 2018, 09:30 AM   #6
Senior Member

Joined: May 2015
From: Arlington, VA

Posts: 444
Thanks: 29

Math Focus: Number theory
Quote:
 Originally Posted by skipjack Interesting question! Is each circle's area the area of the smaller part of the sphere's surface that is enclosed by the circle? If so, the answer must be less than 4pi (the surface area of the sphere), and so is at most 12. I would imagine that 7 non-overlapping circles could be accommodated quite easily. I think the answer is 10.
skipjack seems to have interpreted my description correctly

 June 24th, 2018, 10:02 AM #7 Senior Member   Joined: May 2015 From: Arlington, VA Posts: 444 Thanks: 29 Math Focus: Number theory My general question might loosely be stated as "What is the maximum number of circles of area one which do not overlap, and pack onto a spherical surface of radius N, where N is a nonzero integer?" My correspondent says: "how do you define the circle area? Is it the area of a "flat" circle with its circumference on the surface of the sphere, or is it the (larger) curved area of a spherical cap bounded by a circle on the sphere surface? Of course, the difference between the two definitions becomes negligible for large sphere radii, but for the small values of N the results are different. In general, it seems that you are looking for the results of the "Tammes" problem, for which the best known, most likely optimal solutions up to 130 circles are tabulated on Neil Sloane's web page http://neilsloane.com/packings/index.html#I (results for D=3) http://neilsloane.com/packings/dim3/ For larger sphere radii (larger already means N>3), good approximations would be the packings from spherical codes with icosahedral symmetry http://neilsloane.com/icosahedral.codes/index.html The achievable maximum density of circles on the sphere surface would asymptotically reach the well known Pi*sqrt(3)/6 ~= 0.9069 of the hexagonal lattice packing of circles in the plane. If you are you interested in 'exact' values, then the only proven results are for N=1 (9 or 10 circles dependent on the area definition), and unproved results with good numerical evidence would be 40 or 42 circles for N=2 and 94 or 95 circles for N=3." SDK: Please see that you were right about the ambiguity of my question regarding area surrounded by a circle on the sphere; skipjack, you estimated well with the guess of 10 circles for N=1.

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post fivestar Calculus 2 January 23rd, 2017 08:45 PM Lucasb44 Algebra 1 November 17th, 2011 01:17 PM Thehelmuth Algebra 1 October 1st, 2011 02:24 PM mydisguise08 Algebra 2 November 25th, 2009 03:36 PM 1210 Algebra 7 September 22nd, 2007 12:45 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top