Question
Asked 4 September 2024

I have a new interesting geometric problem. Do you have any ideas on how to solve it?

Given:
In an isosceles triangle, the lengths of the two equal sides are each 1, and the base of the triangle is m.
A circle is circumscribed around the triangle.
Find the chord of the circle that intersects the two equal sides of the triangle and is divided into three equal segments by the points of intersection.

Most recent answer

Dinu Teodorescu
Valahia University of Targoviste
Alex Ravsky Sorry for not read carefully the last part of your comment! Yes, in fact a lot of triangles with lenghts 1, 1, m do not admit the second solution! To have 2 solutions, condition 2-2cosA<=1/2 i.e. cosA>=3/4 is necessary! Very restrictive and cosA>=3/4 an weird limitation for angle A.
Liudmyla Hetmanenko A nice and interesting problem!
3 Recommendations

Popular answers (1)

Alex Ravsky
Pidstryhach Institute for Applied Problems of Mechanics and Mathematics
Людмила, доброго дня.
Я випадково побачив Ваше питання.
Воно просте.
Let x=AK and y=MK=KT=TN.
Then x(1-x)=y*2y=AT*(1-AT).
So either AT=x=AK or AT=1-x.
Suppose first that AT=AK.
Then the triangles ABC and ATK are similar, so y=KT=m*AT.
Therefore 2y^2=(y/m)(1-y/m), 2ym^2=m-y, and y=m/(1+2m^2).
Suppose now that AT=1-x.
Then y^2=KT^2=
AK^2+AT^2-2*AK*AT*cos A=
x^2+(1-x)^2-2x(1-x)cos A=
2x^2-2x+1-4y^2cos A=
2x(x-1)+1-4y^2cos A=
-4y^2+1-4y^2cos A.
So y^2=1/(5+4cos A).
We have m^2=1^2+1^2-2cos A, so 2cos A=2-m^2.
Therefore y^2=1/(9-2m^2).
But we have 1/4\ge x(1-x)=2y^2, so in this case
y^2\le 1/8, that is m\le 1/sqrt{2}.
3 Recommendations

All Answers (6)

Dinu Teodorescu
Valahia University of Targoviste
R denotes the radius of circle(ABC) and MK=KT=TN=y.
Using power of point K with respect to the circle we have
KMxKN=R^2-OK^2 => R^2-OK^2=2y^2.
Now using power of point T with respect to the circle we have
TMxTN=R^2-OT^2 => R^2-OT^2=2y^2
Therefore OK=OT.
By Heron's formula we obtain the area S of triangle ABC
S^2=(m/2+1)(1-m/2)(m/2)(m/2) => S=((1/4)m)sqrt(4-m^2) =>R=1/sqrt(4-m^2).
Now we must find OK, but at the moment I don't have any idea!
1 Recommendation
Alex Ravsky
Pidstryhach Institute for Applied Problems of Mechanics and Mathematics
Людмила, доброго дня.
Я випадково побачив Ваше питання.
Воно просте.
Let x=AK and y=MK=KT=TN.
Then x(1-x)=y*2y=AT*(1-AT).
So either AT=x=AK or AT=1-x.
Suppose first that AT=AK.
Then the triangles ABC and ATK are similar, so y=KT=m*AT.
Therefore 2y^2=(y/m)(1-y/m), 2ym^2=m-y, and y=m/(1+2m^2).
Suppose now that AT=1-x.
Then y^2=KT^2=
AK^2+AT^2-2*AK*AT*cos A=
x^2+(1-x)^2-2x(1-x)cos A=
2x^2-2x+1-4y^2cos A=
2x(x-1)+1-4y^2cos A=
-4y^2+1-4y^2cos A.
So y^2=1/(5+4cos A).
We have m^2=1^2+1^2-2cos A, so 2cos A=2-m^2.
Therefore y^2=1/(9-2m^2).
But we have 1/4\ge x(1-x)=2y^2, so in this case
y^2\le 1/8, that is m\le 1/sqrt{2}.
3 Recommendations
Dinu Teodorescu
Valahia University of Targoviste
Dear Alex, Dear Liudmyla,
However something is weird here! Why we don't have unicity? How this fact can be geometrically explained ?
For example in the case of an equilateral triangle ABC( m=1) !?
Dinu Teodorescu
Valahia University of Targoviste
Alex Ravsky Liudmyla Hetmanenko Of course, from AK(1-AK) = 2y^2=AT(1-AT) it results AK=AT or AK=1-AT, but two valid cases for MN divided in 3 parts by points K on AC and T on AB in the case m=1( ABC equilateral triangle) seems to be weird !
Alex Ravsky
Pidstryhach Institute for Applied Problems of Mechanics and Mathematics
Dinu Teodorescu, right, two different constructions of the chord MN came from two different solutions AK=AT or AK=1-AT of the equation AK(1-AK)=AT(1-AT). But for the equilateral triangle we have only the first solution, because the second solution requires m\le 1/sqrt{2}, see the end of my answer.
2 Recommendations
Dinu Teodorescu
Valahia University of Targoviste
Alex Ravsky Sorry for not read carefully the last part of your comment! Yes, in fact a lot of triangles with lenghts 1, 1, m do not admit the second solution! To have 2 solutions, condition 2-2cosA<=1/2 i.e. cosA>=3/4 is necessary! Very restrictive and cosA>=3/4 an weird limitation for angle A.
Liudmyla Hetmanenko A nice and interesting problem!
3 Recommendations

Similar questions and discussions

Related Publications

Article
Full-text available
Skills in working on the geometry problems great needs of the competence of Geometric Reasoning. As a teacher candidate, State Islamic University (UIN) students need to have the competence of this Geometric Reasoning. When the geometric reasoning in solving of geometry problems has grown well, it is expected the students are able to write their ide...
Article
Presents Heron's original geometric proof to his formula to calculate the area of a triangle. Attempts to improve on this proof by supplying a chain of reasoning that leads quickly from premises to the conclusion. (MDH)
Chapter
This chapter presents Euclidean approach as a learning trajectory to create a common artefact used in Islamic art. The Realistic Mathematics Education sets the theoretical framework for the learning trajectory in two stages, namely, horizontal and vertical mathematization. The chapter provides a brief literature review to explain this framework and...
Got a technical question?
Get high-quality answers from experts.