The Student Room Group

A level maths question about tangents

I was wondering if you can always set the discriminant=0 for a tangent to a curve if you have a quadratic in x. surely you cant always do this since you can have a tangent to a curve at one point which then intersects with the curve at another point so there would be 2 solutions for x, no?
Reply 1
Original post by FM1/FP1
I was wondering if you can always set the discriminant=0 for a tangent to a curve if you have a quadratic in x. surely you cant always do this since you can have a tangent to a curve at one point which then intersects with the curve at another point so there would be 2 solutions for x, no?

How can a tangent to a quadratic intersect the curve again? You might need to show a picture of what you mean.
Original post by FM1/FP1
I was wondering if you can always set the discriminant=0 for a tangent to a curve if you have a quadratic in x. surely you cant always do this since you can have a tangent to a curve at one point which then intersects with the curve at another point so there would be 2 solutions for x, no?


A tangent can't intersect a quadratic curve, because that is a line that would go through two points of the curve. That is not the case for a tangent. A tangent touches the curve on a certain point, that's it. But a secant can do this.

Would be nice, when you send the question paper to it for a better understanding of the excercise.
(edited 1 year ago)
Reply 3
Original post by Notnek
How can a tangent to a quadratic intersect the curve again? You might need to show a picture of what you mean.


Original post by Kallisto
A tangent can't intersect a quadratic curve, because that is a line that would go through two points of the curve. That is not the case for a tangent. A tangent touches the curve on a certain point, that's it. But a secant can do this.

Would be nice, when you send the question paper to it for a better understanding of the excercise.

I dont necessarily mean a quadratic curve but moreso just a curve/function which you can get into the form of a quadratic when finding it's intersection with a line.

The reason I asked was because i was looking at a maths question where you had a hyperbola and needed to simplify the equation of a tangent and the solution back plugged y=mx+c into the general hyperbola question, expanded, then set the descriminant=0
Original post by FM1/FP1
I dont necessarily mean a quadratic curve but moreso just a curve/function which you can get into the form of a quadratic when finding it's intersection with a line.

The reason I asked was because i was looking at a maths question where you had a hyperbola and needed to simplify the equation of a tangent and the solution back plugged y=mx+c into the general hyperbola question, expanded, then set the descriminant=0


Best to post the actual question/solution so as to be clear what you're talking about.

But guessing: A tangent to an hyperbola will only touch the curve once (and not intersect/touch it again). So, if the x-coordinates of the point of intersection are the subject of a quadratic equation, then for a tangent, it can have only one, repeated root , and hence the discriminant must be 0.

Keypoints: The tangent can only touch/intersect the curve once, and you get a quadratic for its x-coordinate. Think this is true of any conic section. Edit: Certainly for a hyperbola or parabola.
(edited 1 year ago)
Original post by ghostwalker
Best to post the actual question/solution so as to be clear what you're talking about.

But guessing: A tangent to an hyperbola will only touch the curve once (and not intersect/touch it again). So, if the x-coordinates of the point of intersection are the subject of a quadratic equation, then for a tangent, it can have only one, repeated root , and hence the discriminant must be 0.

Keypoints: The tangent can only touch/intersect the curve once, and you get a quadratic for its x-coordinate. Think this is true of any conic section.

Agree. Thinking about the asymptotes make it reaonably clear.
https://plus.maths.org/content/meet-hyperbola
Reply 6
Original post by mqb2766
Agree. Thinking about the asymptotes make it reaonably clear.
https://plus.maths.org/content/meet-hyperbola


ahhhhhhhhhhhh, this makes alot of sense, the tangent will be above/below the asymptote so will never reconnect with the curve. thank you
Reply 7
Original post by ghostwalker
Best to post the actual question/solution so as to be clear what you're talking about.

But guessing: A tangent to an hyperbola will only touch the curve once (and not intersect/touch it again). So, if the x-coordinates of the point of intersection are the subject of a quadratic equation, then for a tangent, it can have only one, repeated root , and hence the discriminant must be 0.

Keypoints: The tangent can only touch/intersect the curve once, and you get a quadratic for its x-coordinate. Think this is true of any conic section. Edit: Certainly for a hyperbola or parabola.


image_2023-03-29_135713870.png
The fact that a tangent only intersects with the curve once answers what I didnt understand, lol you answered my question without even seeing it, thank you 😁
To expand a bit on the why:

The big idea is you follow your normal procedure of finding intersecting points of a curve and a line.
If you plug in your equation of tangent (I don't know what it is, but it will be of degree 1, i.e. some sort of y=mx+c) into your quadratic in whatever shape (of degree 2)*, the result is an equation of degree 2, i.e. it has exactly 2 roots**.
The point of contact is a repeated root - it counts as 2. That uses up all the "quota" for the roots of the equation.

*All the conics! Circles, parabolas, hyperbolas, ellipses, you name it.
**That's called Fundamental Theorem of Algebra. Intuitive result, unexpectedly hard to prove. See you in complex analysis at uni lol.


Ref:
https://math.stackexchange.com/questions/4356957/why-tangents-to-a-quadratic-curve-never-cut-it-again
A little discussion on MathStackExchange regarding your question.
(edited 1 year ago)

Quick Reply

Latest