Placing a pole with maximal radius subject to constraints

Click For Summary
The discussion revolves around optimizing the placement of a smaller circle within a defined area constrained by a larger circle and a square. The user seeks to maximize the radius of the smaller circle while adhering to specific geometric constraints involving angles and distances from lines. They express uncertainty about the existence of a closed-form solution and inquire about optimization tools or alternative solving methods. A key insight is that the smaller circle will touch both the edge of the square and the arm at a 45-degree angle. The thread highlights the mathematical relationships and calculations involved in determining the optimal radius.
GuyWhoOnceToldYou
Messages
2
Reaction score
0
TL;DR
Looking to place a pole with maximal radius, constrained by objects surrounding it
I have a problem that I imagine does not have a closed-form solution and requires the use of some kind of optimization solver. I am not an engineer myself, so forgive me if the question seems stupid.

The problem is as follows: I have a circle bound in a square, and an arm going from the center of the circle to the corner of the square (so, 45 degrees), and I need to place a circle with the largest radius possible in the area bound between the outside of the big circle, the inside of the square, and below the arm (the area in green in the image)
1718302190798.png


I can describe these constraints mathematically: I want to maximize some r subject to the constraints: let 2*w be the total width of the arm, and r_circle be the radius of the black circle, then the constraints are:

(r+r_circle)*(cos(alpha),sin(alpha)) is at least distance w+r from the line y=x (I can directly calculate the closest point on the line as a function of r and alpha to simplify this one, I just haven't done that yet).
(r+r_circle)*(cos(alpha),sin(alpha)) is at least distance r from the line from the line x=r_circle (the right edge of the square) (again, I can calculate the point as a function of r, alpha directly)
alpha is between 0 and 45 degrees


So what I'm looking for is either (A) a better way to solve this, or (B) a tool that can solve this.

Thanks in advance!
 
Engineering news on Phys.org
Where the big circle has radius R;
The small circle will have a radius; r = R * 0.123308641756286

R = 1.
Small circle centred at x = 1 - r ; y = x - r * Sqrt(2)
Distance between centres d = Sqrt( x*x + y*y ) ;
Find r, for d - r = 1 ;
 
  • Like
Likes DeBangis21 and GuyWhoOnceToldYou
Baluncore said:
Small circle centred at x = 1 - r ; y = x - r * Sqrt(2)
Of course! I can assume the resulting circle will touch both x=1 and the arm at 45 degrees, that didn't occur to me. Thank you! I will have to adjust to account for the width of the arm, but I think that would be easy.
 
Welcome to PF.

0.123308641756286 = 1 / ( 3 + √2 + 2*√( 2 + √2 ) )
0.123308641756286 = 5 - 3*√2 - 2*√( 10 - 7*√2 )
 
Here's a video by “driving 4 answers” who seems to me to be well versed on the details of Internal Combustion engines. The video does cover something that's a bit shrouded in 'conspiracy theory', and he touches on that, but of course for phys.org, I'm only interested in the actual science involved. He analyzes the claim of achieving 100 mpg with a 427 cubic inch V8 1970 Ford Galaxy in 1977. Only the fuel supply system was modified. I was surprised that he feels the claim could have been...