Hello it would be great if you could help with the following problem.

I had to solve it geometrically (no problem there) and then I thought I'd like to solve it by calculating the points as well. Well that wasn't so easy so here's the problem:


  1. a circle $k: x^2+y^2=6^2$

  2. a point $P = (4,-3.5)$

  3. the distance $\overline{AB}=7$

Now find the isosceles triangle ABC where $A,B,C \in k$ and P lies on AB. There are two solutions for this problem. I already managed to use the parameters to get from 4 equations to 2 equations with 2 unknowns. But not even Mathematica wants to solve those. I can post them here but I didn't want to influence the answers with them yet.

Thanks for any help. :)


Apparently you want me to provide my part of the solution so here we go. I'm lookking for points $A$ and $B$

Let them have the coordinates $A=(x_A,y_A)$ and $B=(x_B,y_B)$

I can use that to get the following 4 equations with 4 variables with the given points:

  1. using the circle: $y_A^2=\pm\sqrt{6^2-x_A^2}$ and $y_B^2=\pm\sqrt{6^2-x_B^2}$

  2. using a line through $A,B,P$: $y_A=\frac{(y_B+3.5)*(x_A-x_B)}{(x_B-4)}$

  3. using the distance: $(x_B-x_A)^2+(y_B-y_A)^2=7^2$

plugging 1. into 2. and 3. I get 2 equations with two unknowns.

  1. $\to$ 2. $y_A=\frac{(y_B+3.5)* \left( \sqrt{6^2-y_A^2}-\sqrt{6^2-y_B^2 }\right)}{\left(\sqrt{6^2-y_B^2} - 4\right)}$


  1. $\to$ 3. $\left(\sqrt{6^2-y_B^2} - \sqrt{6^2-y_A^2} \right)^2 + (y_B-y_A)^2=7^2$

    That's what I got and that doesn't seem like a sensible line of questioning.

  • $\begingroup$ The point $(4,-3.5)$ doesn't seem to be on the circle. $\endgroup$ – zoli Mar 22 '17 at 15:48
  • $\begingroup$ @zoli Fortunately not. Otherwise it would be difficult to find a chord through it with a length of 7. $\endgroup$ – Reinhard Meier Mar 22 '17 at 16:41
  • $\begingroup$ @ReinhardMeier: Oh, yes. $\endgroup$ – zoli Mar 22 '17 at 16:44
  • $\begingroup$ Is $\overline{AC}$ meant to be one of the two equal sides? $\endgroup$ – amd Mar 23 '17 at 0:17
  • $\begingroup$ Either way, aren’t there at least four solutions? There are two chords of length 7 through $P$ and for each chord there are either two of four isosceles triangles that can be built on to (depending on the answer to my previous question). $\endgroup$ – amd Mar 23 '17 at 2:23

You can calculate everything that is constructable with compass and straightedge. So you can try to find a way to get your points $A$, $B$ and $C$ with compass and straightedge and then try to transform this into calculation.

This is one possible approach:

Do a drawing at the side with a triangle $A'B'M'$, where $|A'B'|=7$, $|A'M'|=|B'M'|=6$. Now draw a circle with the radius $|MP|=\sqrt{28.75}$ around the center $M'$. This circle will intersect $A'B'$ at two points $P'_1$ and $P'_2$. The distance $|P'_iA'|$ is a candidate for the distance $|PA|$ and the distance $|P'_iB'|$ is a candidate for the distance $|PB|$. Draw a circle with radius $|P'_1A'|=|P'_2B'|$ around $P$ and a circle with radius $|P'_1B'|=|P'_2A'|$ around $P$. Each of these circles will intersect the original circle at two points. One point on the smaller circle around $P$ together with one point on the bigger circle around $P$ forms a chord through $P$ with length $7$. Once you have $A$ and $B$, it is easy to find $C$ such that you get an isosceles triangle.

Note: I have started to do some of the calculation, e.g. I have found $$ |P'_1A'|=|P'_2B'| = \frac{7}{2} - \frac{3}{2}\sqrt{2} \\ |P'_1B'|=|P'_2A'| = \frac{7}{2} + \frac{3}{2}\sqrt{2} $$ But the number grew more and more ugly and finally I decided not to continue on this.

  • $\begingroup$ thanks for the answer. Ii edited my question to show my try for a solution. can you tell me why that didn't work? thanks $\endgroup$ – Notgiven Mar 24 '17 at 9:08
  • $\begingroup$ @Notgiven There is no particular step in your calculation where I could point to and say this is where you did something wrong. It is not unusual that some attempts to solve a mathematical problem turn out to be dead ends. Sometimes it is a matter of luck, sometimes a matter of experience, sometimes a matter of intuition and sometimes a matter of effort to find an adequate way to solve a problem. By back luck, you have chosen an approach that did not take you anywhere. You could have started with a completely different approach, e.g. trying to avoid the square roots as long as possible. $\endgroup$ – Reinhard Meier Mar 24 '17 at 11:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.