|
||
Title: William the Logger Post by rloginunix on Apr 1st, 2015, 6:43pm William the Logger. Down the Math River that has parallel banks and a width a Willy Wu is floating logs which can be approximated as absolutely rigid line segments. At some point the river makes a ninety-degree turn and while its banks remain parallel its width is now b: http://www.ocf.berkeley.edu/~wwu/YaBBAttachments/rlu_wtl.png For a given a and b what is the longest log L can William safely float around the bend in such a way that the log does not get bent, broken, stuck, submerged below or popped above water (when only 2-dimensional transformations are allowed)? Put it another way. If William knows a, b and L ahead of time when should he make the trip? |
||
Title: Re: William the Logger Post by jollytall on Apr 1st, 2015, 9:45pm First idea: Make the log in alpha degree compared to the horizontal line, when it is almost stuck. Then the "corner" splits it into two parts, x across river a and y across river b. x*cos(alpha) = a y*sin(alpha) = b The log is a/cos(alpha)+b/sin(alpha) The minimum of it is when the derivative is 0, i.e. -a*cos(alpha)/sin2(alpha)+b*sin(alpha)/cos2(alpha)=0. tg(alpha) = (a/b)^(1/3) From alpha x+y = L can be calculated. |
||
Title: Re: William the Logger Post by pex on Apr 1st, 2015, 10:06pm on 04/01/15 at 21:45:31, jollytall wrote:
That's pretty much what I did. The result is rather cute: L = [hide](a2/3 + b2/3) 3/2[/hide]. |
||
Title: Re: William the Logger Post by rloginunix on Apr 2nd, 2015, 9:28am You guys are both correct and what is rather amazing is that I took exactly the same route. Interesting. There is also a more analytic (academic) approach, if you will: [hide]the orthogonal banks are actually the XOY axes. The log must slide along both playing a role of a tangent to some curve. So the question is - find the curve the length of tangent to which trapped between the axes is constant at all times. If the (a,b) corner is inside that curve - no go. Pythagoras and an equation of a straight line lead to a differential equation[/hide] in this case. Turns out that the equation given by pex describes a [hide]curve known as astroid[/hide]. Well done. (I interpreted jollytall's last equation like so: [hide]tg(http://www.ocf.berkeley.edu/~wwu/YaBBImages/symbols/alpha.gif) is almost a derivative, for it to be a real one the angle must be http://www.ocf.berkeley.edu/~wwu/YaBBImages/symbols/pi.gif- http://www.ocf.berkeley.edu/~wwu/YaBBImages/symbols/alpha.gif and then we get a trivial differential equation dy/dx = -(y/x)1/3[/hide] ...) PS This particular puzzle interconnects with [hide]A Cat on a Ladder[/hide] and [hide]Infinite Fast Ladder[/hide] in the easy section. [edit] Fixed two spelling errors. [/edit] |
||
Powered by YaBB 1 Gold - SP 1.4! Forum software copyright © 2000-2004 Yet another Bulletin Board |