A certain acute isosceles triangle has the same circumradius and inradius as a right triangle with sides 3 , 4 , and 5 . If one of the base angles of that acute isosceles triangle is ∠ P , then sec P = ϕ + n , where ϕ = 2 1 ( 1 + 5 ) , the golden ratio, and n is an integer. Find n .
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.
Without loss of generality, since the right triangle △ A B C (of side lengths 3 , 4 and 5 ) is not isosceles, the circumcenter O 1 does not lie within the incircle of center O 2 , which automatically produces one isosceles acute triangle and one isosceles obtuse triangle both having the congruent inradii and circumradii. Thus, for this problem, we look at △ P N R as the circumcenter O 1 does not lie inside the obtuse isosceles triangle.
By Euler's theorem , since the circumradius and the inradius of △ A B C are respectively R = 2 5 and r = 1 , the distance between the circumcenter and the incenter is
∣ O 1 O 2 ∣ 2 ∣ O 1 O 2 ∣ = R ( R − 2 r ) = 2 5 ( 2 5 − 2 ) = 2 5
which shows that the height of the isosceles triangle is
∣ R M ∣ = ∣ R O 1 ∣ + ∣ O 1 O 2 ∣ + ∣ O 2 M ∣ = 2 5 + 2 5 + 1 = 2 1 ( 7 + 5 )
So since
circumradius of isosceles triangle = 8 1 ( ∣ R M ∣ ∣ P N ∣ 2 + 4 ∣ R M ∣ )
With some algebra, we found that ∣ P N ∣ = 2 4 − 5 . Thus, as 0 ∘ < ∠ P < 9 0 ∘ ,
sec 2 ∠ P sec ∠ P = 1 + tan 2 ∠ P = 1 + tan 2 ∠ P = 1 + tan 2 ( arctan ( 2 1 ∣ P N ∣ ∣ R M ∣ ) ) = 2 1 ( 5 + 5 ) = 2 1 ( 1 + 5 ) + 2
where the answer to this problem is n = 2 .
If suppose we are interested in a more algebraic approach (assuming we don't have some tool to bypass by geometry), we can simply apply the following formulas:
circumradius of isosceles triangle inradius of isosceles triangle = 8 1 ( h a 2 + 4 h ) = 4 h a ( a 2 + 4 h 2 − a )
where a is the side length opposite from the apex vertex, and h is the height. In terms of ( a , h ) , the solutions are S 1 = ( 2 4 − 5 , 2 1 ( 7 + 5 ) ) and S 2 = ( 2 4 + 5 , 2 1 ( 7 − 5 ) ) . But since 2 1 ( 7 − 5 ) < 2 5 , S 2 gives the obtuse triangle. So the rest of the steps involve using S 1 to find ∠ P . This result agrees with my solution that inspects all possible triangles with less algebra.
I might have missed something, but is the proof of the above construction obvious?
Extending O 2 O 1 gives us point R , and drawing tangents from R to the incircle gives us the isosceles △ P R N . But why is P N tangent to the incircle ?
Log in to reply
That is because of the properties of circumcircles and incircles. What makes the construction "obvious" is that the height of isosceles triangles must be overlapped by the diameter of the circumcircle. One can easily impose symmetry as shown below, which gives the following illustration to the solution:
From what we have seen, as the inradius depends on the right triangle (with the circumradius fixed), it is not likely that the side opposite to the apex point would ever not be tangent to the incircle.
Side Note: One can generalize that from using the inradius and circumradius formulas. If suppose we set r = 1 , the inradius of the right triangle is bounded by sin θ + cos θ + 1 sin θ cos θ . As that leads to solving the polynomial (with change of variables), we found two real-valued solutions.
Log in to reply
I have posted the follow-up problem , which extends what was being discussed here. David, the problem you posted is golden as there is some good driving point!
Fun fact: if you work out the obtuse isosceles triangle with the same inradius and circumradius instead (which possibly I did by mistake), sec P = 2 − 2 5 − 1 and the length of the shorter sides of the triangle happens to be 3 . 4 5 … .
We'll use the identity cos(A)+cos(B)+cos(C)=1+(r/R) where A,B,C are angles of triangle and r and R are the inradius and circumradius respectively. We can easily calculate that r=1,R=5/2 and takeA=B=P. Thus we have 2cos(P)-cos(2P)=7/5 => 2cos²(P)-2cos(P)+2/5=0 => 5cos²(P)-5cos(P)+1=0 =>t²-5t+5=0....taking sec(P)=t =>x²-x-1=0.....taking t-2=x => sec(P)=x+2. Now we the above quadratic has root of the golden ratio...thus sec(P)=x+2, hence n=2 is our answer.
A nice and clean approach! Thanks for sharing!
Thank you David :)
The circumradius of a triangle is R = 4 T a b c , where a , b , and c are the sides of a triangle and T is its area. For the right triangle with sides 3 , 4 , and 5 , a = 3 , b = 4 , c = 5 , and T = 2 1 ⋅ 3 ⋅ 4 = 6 , so R = 4 T a b c = 4 ⋅ 6 3 ⋅ 4 ⋅ 5 = 2 5 .
The inradius of a triangle is r = a + b + c 2 T , where a , b , and c are the sides of a triangle and T is its area. For the right triangle with sides 3 , 4 , and 5 , a = 3 , b = 4 , c = 5 , and T = 2 1 ⋅ 3 ⋅ 4 = 6 , so r = a + b + c 2 T = 3 + 4 + 5 2 ⋅ 6 = 1 .
Let the base of the acute isosceles triangle be 2 x , its legs be y , and its height be h .
Then the sides of the acute isosceles triangle are a = 2 x , b = y , c = y , and T = 2 1 ⋅ 2 x ⋅ h = x h , so R = 4 T a b c = 4 x h 2 x ⋅ y ⋅ y = 2 h y 2 = 2 5 , which means y 2 = 5 h .
and r = a + b + c 2 T = 2 x + y + y 2 x h = x + y x h = 1 , which means x h = x + y .
By the Pythagorean Theorem, x 2 + h 2 = y 2 . These three equations solve to ( h , x , y ) = ( 2 1 ( 7 + 5 ) , 4 − 5 , 2 1 4 − 5 ( 5 + 5 ) ) or ( h , x , y ) = ( 2 1 ( 7 − 5 ) , 4 + 5 , 2 1 4 + 5 ( 5 − 5 ) ) , but only ( h , x , y ) = ( 2 1 ( 7 + 5 ) , 4 − 5 , 2 1 4 − 5 ( 5 + 5 ) ) leads to an acute isosceles triangle.
For this acute isosceles triangle, sec P = x y = 4 − 5 2 1 4 − 5 ( 5 + 5 ) = 2 1 ( 5 + 5 ) = ϕ + 2 for ϕ = 2 1 ( 1 + 5 ) . Therefore, n = 2 .
Problem Loading...
Note Loading...
Set Loading...
The inradius r of a triangle with sides a , b , c and area A is r = a + b + c 2 A , thus the inradius of the 3 - 4 - 5 right triangle is r = 3 + 4 + 5 2 × 2 1 × 3 × 4 = 1 .
The hypotenuse of a right triangle is a diameter of its circumcircle, thus the circumradius of the 3 - 4 - 5 right triangle is R = 2 5 .
Let O be the circumcenter and I the incenter of our isosceles △ T P Q , ( T P = T Q ) and D , E the intersection points of the incircle with its sides P Q , T P , as seen in the figures. By Euler's theorem the distance between the circumcenter and the incenter is d = R ( R − 2 r ) , hence O I = 2 5 ( 2 5 − 2 × 1 ) = 2 5 This means that there are two ways point I can be placed on the axis of symmetry of the isosceles △ T P Q : I may lie either inside segment O T , or outside it. The first case occurs when △ T P Q is an obtuse triangle, the second when it is an acute one. This can be verified by showing that ∠ I T E is greater than, or less than 4 5 ∘ respectively.
Indeed, on △ I T E , in the first case (figure 1) we have sin 2 T = I T I E = R − d r = 2 5 − 2 5 1 = 1 0 5 + 5 > 2 2 = sin 4 5 ∘ Since 0 ∘ < 2 T < 9 0 ∘ , we get that 2 T > 4 5 ∘ , thus △ T P Q is an obtuse triangle.
In the second case (figure 2), sin 2 T = I T I E = R + d r = 2 5 + 2 5 1 < 2 2 thus △ T P Q is an acute triangle.
In the problem we are interested in the latter case. But then, on right △ T P D , sec P = csc 2 T = 2 5 + 2 5 = 2 1 + 5 + 2 For the answer, n = 2 .