I was wondering how a curve would look like if you have two hands of a clock moving at the same speed from different positions while you track the x of one hand and the y of the other hand over time. That's how I got (cos(t),sin(t+a)) where t is the time and a the angle between the hands.
It turned out to almost always describe an ellipse tilted by 45°. I have no idea why it is always tilted by that angle and how to show this parametrisation describes an ellipse.
My failed tries: 1. Rotating (cos(t),sin(t+a)) by 45° and showing that the rotated points suffice the $\frac{x^2}{u^2}+\frac{y^2}{v^2}=1$ equation.
- Using the distance to focal points definition for (cos(t),sin(t+a))
Maybe it's not possible to show in general because of the special cases where (cos(t),sin(t+a)) describes a line. Please help!
I also failed showing it for a special a.
Sry for my bad english. I'm no native speaker
Eliminate $t$ by expanding $y=\sin(t+a)$ into $y=\cos t\sin a+\cos a\sin t$ and then solve for $\sin t$ to get $(y-x\sin a)\sec a$, so that $$x^2+(y-x\sin a)^2\sec^2a=1,$$ which can be rearranged into the more standard-looking $$x^2-2xy\sin a+y^2=\cos^2a. \tag{*}$$ For $\cos a\ne0$, this equation indeed describes an ellipse. Using any of the usual methods, such as orthogonal diagonalization, its principal axes can be found to be in the directions $(1,1)$ and $(-1,1)$, i.e., its major axis is rotated 45° either clockwise or counterclockwise from standard position. When $\sin a=0$, you have a circle.
When $\cos a=0$, the implicit Cartesian equation (*) becomes $(x\pm y)^2=0$, i.e., a double line, but the original parametric equation only produces values of $x$ and $y$ in the interval $[-1,1]$, so the ellipse collapses to a line segment in this case.