This is a method I tried of working out the cartesian form of a hyperbola using its definition as the cross section of a vertical cone?
I started by noting that the width of the hyperbola is the sine of a circle at some point within the cone. The radius of the circle is proportional to the height of the cone, which is the length $y$ down the hyperbola plus some constant $a$. Therefore the radius of the circle is $k(y+a)$.
The centre of the hyperbola is always the same distance from the centre of the cone at a given height $y$, at $ka$. So cosine of the angle between the middle and edge of the hyperbola at some height $y$ is $\frac{ka}{k(y+a)} = \frac{1}{1+\frac{y}{a}}$. So the width of the hyperbola $x$ at height $y$ is $x =k(y+a)\sqrt{1-\frac{1}{(1+\frac{y}{a})^2}}$ by relating the sines to cosines. The resulting equation is $x^2 = k^2(y+a)^2(1-\frac{1}{(1+\frac{y}{a})^2})$,
This simplifies to $x^2 = k^2(y^2 + 2ya)$. Is this a hyperbolic curve. How do I rearrange it to the standard form?
I understand that I'm not being very clear but I can't find a software on which to draw the problem properly. Anyway, I hope that the line of reasoning has come across.
$-\frac{x^2}{k^2} + (y+a)^2 = a^2$
$-\frac{x^2}{a^2k^2} + \frac{(y+a)^2}{a^2} = 1$
This is the equation of a hyperbola on the vertical transverse axis with vertex $a$, asymptote, $y=\frac{1}{k}x$ and translated down by $a$, indeed a correct parametrisation for the curve of the hyperbola