Triangle A O B and triangle B O C are directly similar (they have the same orientation). Let the centroids of these two triangles be M 1 and M 2 , respectively.
If triangle M 1 O M 2 is right, and M 1 M 2 = A B , then the sum of all possible values of cos 2 ( ∠ A B O ) can be expressed as q p for positive coprime integers p , q .
Find p + q .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
@Calvin Lin Can you please let all the people who answered 2 3 be correct? I accidentally typoed cos and sin .
Log in to reply
@Calvin Lin tagged above
Those who previously entered 23 have been marked correct.
Log in to reply
Apparently nobody answered 23 though, so no harm done I guess.
For the lemma, it directly follows from this observation:
Let
T
be the transformation that rotates about
O
by
∠
A
O
B
, and scales by
O
A
O
B
.
All triangles with vertices
X
,
O
,
T
(
X
)
will have a ratio of sides
O
A
O
B
and a internal angle of
∠
A
O
B
, thus they will all be similar.
Since we have
T
(
A
)
=
B
,
T
(
B
)
=
C
,
T
(
M
1
)
=
M
2
and
T
(
O
)
=
0
, hence the triangles
A
O
B
,
B
O
C
,
M
1
O
M
2
are similar.
This in essence is what you're trying to say. The above is a simpler way to motivate the lemma.
Log in to reply
It could be "explained" even simpler by citing the Mean Geometry Theorem, which gives △ M 1 O M 2 = 3 1 △ A O B + 3 1 △ B O C + 3 1 △ O O O which proves the triangles' similarity.
I might create a Wiki article on it, if I find the time.
In triangle M 1 O M 2 , I assumed that the right angle was at vertex M 2 because that's what the diagram indicated, which it turns out is the one case that can't happen. This seems misleading, and I think you should specifically say in the problem that the right angle can be at any vertex.
Log in to reply
Well, I did add "not to scale" on the diagram... I put the right angle at the vertex that can't be a right angle just so I reveal nothing about the solution. Maybe I should specify further that you should NOT make any assumptions based on the diagram?
Log in to reply
Yes, but saying "not to scale" only means you can't assume that the general proportions of the diagram are accurate. But marking a right angle is definitive - you should only do this is there is actually a right angle where you say it is.
Instead of cautioning the reader to not make any assumptions about the diagram, it would be better not to put features in the diagram that don't actually hold.
90 on the diagram is misleading. Can you please correct it ? .Angle
M
1
M
2
O
may be drawn as acute angle. In fact since I saw that it can not be a right angle, I gave up.
I was not able to draw a scaled sketch. Can you please supply a scaled sketch?
You mention that
△
M
1
O
M
2
≅
△
A
O
B
{
o
r
i
s
i
t
≅
△
B
O
A
?
}
∴
O
M
1
=
O
B
.
Can you please explain how is this possible ? Thanks.
Shouldn't we consider cases where BA/BO=AO/OC or BA=OC?
Problem Loading...
Note Loading...
Set Loading...
Lemma: △ M 1 O M 2 ∼ △ A O B ∼ △ B O C
Proof: let the midpoints of A B and B C be P 1 and P 2 , respectively. Note that △ P 1 O P 2 and △ M 1 O M 2 are related by a homothety with center O and scale 2 3 , so they are similar. Thus it remains to prove that P 1 O P 2 ∼ A O B .
Note that P 1 O and P 2 O are the medians from O of △ A O B and △ B O C respectively. Thus P 2 O P 1 O = B O A O .
Also, note that ∠ P 1 O P 2 = ∠ P 1 O B + ∠ B O P 2 = ∠ P 1 O B + ∠ A O P 1 = ∠ A O B .
Thus, △ P 1 O P 2 ∼ △ A O B and we proved our lemma.
Since △ M 1 O M 2 is right, then that implies that △ A O B and △ B O C are also right.
Since M 1 M 2 = A B , it means △ M 1 O M 2 ≅ △ A O B .
Now there are two cases: ∠ A O B = 9 0 ∘ or ∠ O A B = 9 0 ∘ , because clearly ∠ A B O = 9 0 ∘ .
Case 1: ∠ A O B = 9 0 ∘
Let M 1 O = A O = x . Note that P 1 O = 2 3 x . Since P 1 A = P 1 B = P 1 O , then A B = 3 x . Finally, by Pythagorean Theorem, B O = 2 x 2 .
This gives cos 2 ∠ A B O = ( 3 2 2 ) 2 = 9 8
Case 2: ∠ O A B = 9 0 ∘
Let M 1 O = A O = x . Note that P 1 O = 2 3 x . By Pythagorean Theorem, P 1 A = 2 5 x so A B = x 5 . By Pythagorean Theorem, B O = x 6 .
This gives cos 2 ∠ A B O = ( 6 5 ) 2 = 6 5
Finally, 9 8 + 6 5 = 1 8 3 1 so our answer is 3 1 + 1 8 = 4 9 .