Normal to a fixed concentric ellipse

  • Thread starter Thread starter utkarshakash
  • Start date Start date
  • Tags Tags
    Ellipse Normal
AI Thread Summary
The discussion focuses on proving that the line MN is always normal to a fixed concentric ellipse when perpendiculars PM and PN are drawn from any point P on the ellipse. The participants explore the equation of line MN and compare it to the generic equation for a normal to an ellipse, emphasizing the need to express parameters a', b', and φ in terms of a, b, and θ. They highlight the importance of keeping a' and b' fixed while allowing φ to vary with θ to ensure all values can occur without gaps in the ellipses. The conversation involves deriving relationships involving trigonometric functions to establish these conditions. The goal is to demonstrate the geometric properties of the ellipse through algebraic manipulation.
utkarshakash
Gold Member
Messages
852
Reaction score
13

Homework Statement


PM and PN are perpendiculars upon the axes from any point 'P' on the ellipse. Prove that MN is always normal to a fixed concentric ellipse

Homework Equations



The Attempt at a Solution


I assume point P to be (acosθ, bsinθ)

The eqn of line MN is then given by
bsin\theta x+acos\theta y =absin\theta cos\theta
 
Physics news on Phys.org
Try writing the generic equation for a normal to an ellipse with same centre and axes.
 
haruspex said:
Try writing the generic equation for a normal to an ellipse with same centre and axes.

a'sec \phi x - b'cosec \phi y = a^2 - b^2
 
utkarshakash said:
a'sec \phi x - b'cosec \phi y = a^2 - b^2
I assume you meant a'sec \phi x - b'cosec \phi y = a'^2 - b'^2
It remains to find expressions for a', b' and phi in terms of a, b and θ that make this the same as the equation for MN. There is a constraint regarding which of a', b' and phi can depend on which of a, b and θ.
 
haruspex said:
I assume you meant a'sec \phi x - b'cosec \phi y = a'^2 - b'^2
It remains to find expressions for a', b' and phi in terms of a, b and θ that make this the same as the equation for MN. There is a constraint regarding which of a', b' and phi can depend on which of a, b and θ.

Do you want me to compare the two lines?
 
utkarshakash said:
Do you want me to compare the two lines?
Yes. You want to make a′x sec(ϕ)−b′y cosec(ϕ)=a′2−b′2 look like bx sin(θ)+ay cos(θ)=ab sinθ cosθ by suitable choices of a', b' and ϕ. But note that this must work keeping a' and b' fixed while ϕ is allowed to vary as a function of θ.
 
haruspex said:
Yes. You want to make a′x sec(ϕ)−b′y cosec(ϕ)=a′2−b′2 look like bx sin(θ)+ay cos(θ)=ab sinθ cosθ by suitable choices of a', b' and ϕ. But note that this must work keeping a' and b' fixed while ϕ is allowed to vary as a function of θ.

OK I did exactly what you said and got the following relations after comparison

cos\alpha = \dfrac{aa&#039;cos\theta}{a&#039;^2 - b&#039;^2} \\<br /> <br /> sin \alpha = \dfrac{-bb&#039;sin\theta}{a&#039;^2 - b&#039;^2} \\ <br /> <br /> tan \alpha = \dfrac{-bb&#039; tan\theta}{aa&#039;}<br /> <br />
 
utkarshakash said:
OK I did exactly what you said and got the following relations after comparison

cos\alpha = \dfrac{aa&#039;cos\theta}{a&#039;^2 - b&#039;^2} \\<br /> <br /> sin \alpha = \dfrac{-bb&#039;sin\theta}{a&#039;^2 - b&#039;^2} \\ <br /> <br /> tan \alpha = \dfrac{-bb&#039; tan\theta}{aa&#039;}<br /> <br />
OK, nearly there. Now you must set a' and b' so that all values of θ and α can occur. (Otherwise there will be a gap in one of the ellipses.)
 
Back
Top