MedVision ad

Some Conics Questions (1 Viewer)

DcM

Member
Joined
Sep 1, 2003
Messages
53
Gender
Female
HSC
2004
Originally posted by George W. Bush
CM: Is there a easy way to solve the following question without extensive algebra?

The points P and Q lie on an ellipse (standard eqn). The tangents and P and Q intersect at point T. The midpoint of PQ is the point M. Prove that OM extended (where O is origin) passes through the point T.

the way i wud do it is like this...but that seems such an algebra bash...

find midpt of PQ
find eqn of OM
use chord of contact to find out coordinates of T
sub T into eqn to see if it fits..

this way..u have all the weird algebra or trig..so i dunno if it works cos i didn complete doing it...seems too long...

it wud b nice to have a quicker way.......
well CM?
 

DcM

Member
Joined
Sep 1, 2003
Messages
53
Gender
Female
HSC
2004
when M is midpoint of PQ
does that make angle PMT and QMT = 90 degrees?
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
Originally posted by George W. Bush
CM: Is there a easy way to solve the following question without extensive algebra?

The points P and Q lie on an ellipse (standard eqn). The tangents and P and Q intersect at point T. The midpoint of PQ is the point M. Prove that OM extended (where O is origin) passes through the point T.
I've been giving this one some thought - most obvious approaches will result in an algebra bash of unnecessary complexity. I don't see a straight geometric approach, however here is a reasonable algebraic approach.

Let's assume that the point T is (x<sub>0</sub>, y<sub>0</sub>), in which case m<sub>OT</sub> = y<sub>0</sub> / x<sub>0</sub>. Since any two parallel lines with a common point are the same line, it will be sufficient to prove that m<sub>OM</sub> = y<sub>0</sub> / x<sub>0</sub>.

Now, the chord of contact from T is xx<sub>0</sub> / a<sup>2</sup> + yy<sub>0</sub> / b<sup>2</sup> = 1 _____(1)
It meets the ellipse at the points P and Q, and by making y the subject of (1), and putting it into the equation of the ellipse, we get a quadratic equation for the x coordinates of P and Q:
x<sup>2</sup> / a<sup>2</sup> + (1 / b<sup>2</sup>y<sub>0</sub><sup>2</sup>) * (b<sup>2</sup> - b<sup>2</sup>xx<sub>0</sub> / a<sup>2</sup>)<sup>2</sup> = 1
ie x<sup>2</sup>[(1 / a<sup>2</sup>) + (b<sup>2</sup>x<sub>0</sub><sup>2</sup> / a<sup>4</sup>y<sub>0</sub><sup>2</sup>] - 2b<sup>2</sup>xx<sub>0</sub> / a<sup>2</sup>y<sub>0</sub><sup>2</sup> - 1 + b<sup>2</sup> / y<sub>0</sub><sup>2</sup> = 0

We don't care what the solutions are, but we do care what their average is, as it is the x coordinate of M. Thus,
x coordinate of M = -B / 2A = (2b<sup>2</sup>x<sub>0</sub> / a<sup>2</sup>y<sub>0</sub><sup>2</sup>) / 2[(1 / a<sup>2</sup>) + (b<sup>2</sup>x<sub>0</sub><sup>2</sup> / a<sup>4</sup>y<sub>0</sub><sup>2</sup>] = a<sup>2</sup>b<sup>2</sup>x<sub>0</sub> / (a<sup>2</sup>y<sub>0</sub><sup>2</sup> + b<sup>2</sup>x<sub>0</sub><sup>2</sup>)

By making x the subject of (1), we can similarly find an equation whose solution is the y coordinate of P and Q, and use it to show that the y coordinate of M is a<sup>2</sup>b<sup>2</sup>y<sub>0</sub> / (b<sup>2</sup>x<sub>0</sub><sup>2</sup> + a<sup>2</sup>y<sub>0</sub><sup>2</sup>)

Thus, m<sub>OM</sub> = [a<sup>2</sup>b<sup>2</sup>y<sub>0</sub> / (b<sup>2</sup>x<sub>0</sub><sup>2</sup> + a<sup>2</sup>y<sub>0</sub><sup>2</sup>)] / [a<sup>2</sup>b<sup>2</sup>x<sub>0</sub> / (a<sup>2</sup>y<sub>0</sub><sup>2</sup> + b<sup>2</sup>x<sub>0</sub><sup>2</sup>)] = y<sub>0</sub> / x<sub>0</sub>, as required.

Note that this approach requires both x<sub>0</sub> and y<sub>0</sub> to both be non-zero. However, if either of these is zero, then the point T lies on an axes, and symmetry dictates that M must lie on the same axis - try drawing the diagram if you're not convinced - and thus the theorem is obviously true for these cases as well.
 
Last edited:

DcM

Member
Joined
Sep 1, 2003
Messages
53
Gender
Female
HSC
2004
wahhh
that way is sooo much betta..
even tho its still complicated but its a good approach!!

its clever how u used sum of roots to get the midpt..its like one of those tricks and short cuts...i didnt get it at first wen i learnt it @ tutoring...
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
Originally posted by DcM
wahhh
that way is sooo much betta..
even tho its still complicated but its a good approach!!
Thanks - it shows that a bit of thought will usually simplify even an algebra bash Q...
its clever how u used sum of roots to get the midpt..its like one of those tricks and short cuts...i didnt get it at first wen i learnt it @ tutoring...
This is (IMO) a 'standard' trick - always ask yourself if you seek to find something not requested in the question - "Do I really care what this is?" - As an example, consider Q2 in the recent set of conics questions. The best solution involved finding neither PT nor PN - all we cared about was their ratio, and that could be accessed another way. :)
 

OLDMAN

Member
Joined
Feb 20, 2003
Messages
251
Location
Mudgee
Gender
Undisclosed
HSC
N/A
This is a useful thread. Should be bumped up again closer to exam time for revision.
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
Do people want me to post the answers to those questions that have not yet been answered? Alternately, are people still working on them?
 

Calculon

Mohammed was a paedophile
Joined
Feb 15, 2004
Messages
1,743
Gender
Undisclosed
HSC
N/A
The conics tip in this post should be added to the HSC tips forum
 

Grey Council

Legend
Joined
Oct 14, 2003
Messages
1,426
Gender
Male
HSC
2004
for question 4:
I used m<sub>1</sub> * m<sub>2</sub> = -1
but after that I end up using t formulas. :-\

I get b^2/a^2 * sin@sin# / ((cos@-1)(cos#-1))
somehow I have to get the
sin@sin# / ((cos@-1)(cos#-1))
to equal 1.
how?

edit:
that is to say, I CAN do it, but since we aren't looking for algebra bashes, and t formulae are (algebra bashing)^2, me is looking for a better, elegant solution.
 
Last edited:

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top