Community

Looking for a challenging algebra text? Preparing for MATHCOUNTS or the AMC exams?
Check out Art of Problem Solving's Introduction to Algebra by Richard Rusczyk.
Login Register Memberlist Search AoPS Blogs Contests Galleries Forum Index
The time now is Fri Nov 20, 2009 11:54 pm
All times are UTC - 8
View posts since last visit
View unanswered posts
Simson Line Property
Moderators: High School Olympiad Moderators, darij grinberg, N.T.TUAN, orl, pbornsztein, pohoatza, yetti
Post new topic   Reply to topic View previous topicView next topic
8 Posts • Page 1 of 1
Author Message
Michael Niland
Riemann Hypothesis
Riemann Hypothesis

Offline
Joined: 29 Jul 2004
Posts: 460

To rate posts you must be logged in
#1
Simson Line Property
Math Gazette 1918

Prove that the Simson LInes of triangles with the same incircles and circumcircles for a fixed point on the circumference are concurrent.

PostPosted: Sat Mar 19, 2005 8:43 pm  Back to top 
  ProfilePM
yetti
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 31 Dec 2004
Posts: 2108
Location: New York
Czech Republic

To rate posts you must be logged in
#2

me

According to the 2nd Fontene theorem, the orthopole of a line passing through the circumcenter (O) of a triangle \triangle ABC lies on the circumcircle the pedal triangle \triangle XYZ of any point P on this line. For a proof, see Darij Grinberg, Generalization of the Feuerbach point 1). The pedal triangle of the circumcenter O is the medial triangle \triangle A'B'C' and its circumcircle is the 9-point circle (N) of the triangle \triangle ABC. Hence, the orthopoles of all lines passing through the circumcenter lie on the 9-point circle. If the point P is the intersection of a line through the circumcenter with the circuncircle (O), the pedal triangle degenerates into a line segment and its circumcircle into a Simson line with the pole P, which passes through the orthopole of the line OP.
--------------------------------------------------------
The pedal triangle of the incenter I is the contact triangle and its circumcircle is the incircle (I) of the triangle \triangle ABC. The orthopole of the diacentral line OI lies both on the 9-point circle and on the incircle, hence, it is identical with their tangency point F \equiv (I) \cap (N) (Feuerbach point). Let the diacentral line OI intersect the circumcircle and incircle at points P, P' and Q, Q', so that the points P, Q, I, Q', P' and the points P, O, I, P' follow on the line OI in these orders. The orthocenter H is the external homothety center of the circumcircle and the 9-point circle. The Feuerbach point F is the external homothety center of the incircle and the 9-point circle. Thus the line HF is the external homothety axis of the circumcircle, 9-point circle, and the incircle passing through the external homothety center E of the incircle and the circumcircle, the intersection of the diacentral line OI with the homothety axis HF. Let the line HF intersect the circumcircle (O) at a point F'. Because of the homothety between the incircle and circumcircle, the right angle triangles \triangle QFQ' \sim \triangle PF'P' are cenrally similar with homothety cemter E. Because of the homothety between the 9-point circle and circumcircle with homothety center H and homothety coefficient \frac 1 2, the 9-point circle cuts the segments PH, P'H at their midpoints R, R', the lines PP' \equiv QQ' \parallel RR' are parallel and the segment RR' is a diameter of the 9-point circle. Consequently, the right angle triangles \triangle PFP' \sim \triangle RFR' are centrally similar with homothety center H and the right angle triangles \triangle QFQ' \sim \triangle RFR' are centrally similar with homothety center F. The Simson lines of the poles P, P', perpendicular to each other, pass through these midpoints R, R' and they meet at the orthopole F of the diacentral line PP' \equiv OI. Hence, the Simson lines of the poles P, P' are the lines RF, R'F. Because of the homothety of the triangles \triangle QFQ' \sim \triangle RFR', the Simson lines RF, R'F pass through the points Q, Q'. But these 2 points are completely independent of the "rotation" of the triangle \triangle ABC around the fixed circumcircle (O) and the incircle (I) according to Poncelet's porism. As a result, the intersections Q, Q' of the incircle with the diacentral line OI are the fixed points of Simson lines with the poles P, P'.
--------------------------------------------------------
Let P, S be 2 points of the circumcircle with the central angle 0 < \measuredangle POR = \phi \le 180^o. Then the Simson lines s_P, s_S of the poles P, S form an angle of \frac \phi 2. This easily follows from the usual construction of a Simson line in the desired direction. Let the lines parallel to the Simson lines s_P, s_S and passing through one of the triangle vertices, say A, itersect the circumcircle at points K, L different from the vertex A. Thus the angle between the Simson lines s_P, s_S is equal to the angle \measuredangle KAL. The poles P, S are intersections of the lines throught the points K, L perpendicular to the triangle side BC with the circumcircle (O). Hence, the quadrilateral KLSP is a cyclic trapezoid, which means that it is isosceles with KL = PS. Consequently, the central angles \measuredangle KOL = \measuredangle POS = \phi are equal and the angle formed by the 2 Simson lines is equal to \measuredangle KAL = \frac \phi 2. As a special case, if the poles P, S are diametrally opposite points of the circumcircle (O), the angle \phi = \measuredangle POS = 180^o and the Simson lines of the poles P, S are perpendicular to each other.
--------------------------------------------------------
Reflections of any 2 circumcircle diameters PP', SS' in any side line of the medial triangle \triangle A'B'C' meet on the 9-point circle. This is because the circumcenter O of the triangle \triangle ABC is the orthocenter of the medial triangle and the 9-point circle (N) is its circumcircle. The reflection of the orthocenter in any side of a triangle lies on its circumcircle and the reflection of any line passing through the orthocenter in any triangle side must pass through the corresponding orthocenter reflection on the circumcircle. Let the reflections of the diameters PP', SS' in the side line B'C' of the medial triangle meet at a point W on the 9-point circle (N). These reflected lines form the same angle \phi = \measuredangle POS as the diameters PP', SS'. According to the above reference 1), the reflections of the diameters PP', SS' in the sides B'C', C'A', A'B' of the medial triangle concur at the orthopoles U, V of these diameters. Thus the angle \measuredangle UWV spanning the arc UV of the 9-point circle (N) is equal to \measuredangle UWV = \measuredangle POS = \phi and the central angle of the orthopoles U, V on the 9-point circle is equal to \measuredangle UNV = 2 \measuredangle UWV = 2 \phi. As a special case, if the diameters PP', SS' are perpendicular to each other, i.e. \phi = \measuredangle POS = 90^o, their orthopoles U, V are diametrally opposite points of the 9-point circle.
--------------------------------------------------------
Let R, r be the fixed circumradius and inradius, d = OI the fixed distance from the circumcenter to the incenter and \triangle ABC an arbitrary triangle with the fixed circumcircle (O) and incircle (I). The 9-point circle (N) is the circumcircle of the medial triangle \triangle A'B'C', hence, its radius is n = \frac R 2 is also fixed. The 9-point circle is tangent to the fixed incircle at a point F (Feuerbach point). Thus the locus of the 9-point circle centers N of the triangles \triangle ABC "rotated" according to Poncelet's porism is a circle (I, m) centered at the incenter I and with radius m =  n - r = \frac{R - 2r}{2}. Consider the circumcircle diameter PP', which coincides with the diacentral line OI, and an arbitrary circumcircle diameter SS' forming the central angle \phi = \measuredangle POS \le 90^o with the diacentral diameter PP'. If the angle \phi = 0, i.e., if the diameters SS' \equiv PP' are identical, the locus of orthopoles F of this diameter (Feuerbach points) is the incircle (I). If the angle \phi = 90^o, i.e., if the diameters SS' \perp PP' are perpendicular, the orthopole V is the diametrally opposite point of the Feuerbach point on the 9-point circle. Hence, the locus of the orthopoles V diametrally opposite to the Feuerbach points F on the 9-point circles is a circle (I, k) centered at the incenter I and with radius k = m + n = \frac{R - 2r}{2} + \frac R 2 = R - r. If the angle 0 < \phi < 90^o, consider the triangle \triangle INV. The side IN = m = \frac{R - 2r}{2} is fixed, the side NV = n = \frac R 2 is fixed and the angle \measuredangle INV \equiv \measuredangle FNV is also fixed, because the Feuerbach point F lies on the center line IN of the incircle and the 9-point circle and the 9-point circle central angle of the orthopoles F, V of the circumcircle diameters PP', SS' is equal to \measuredangle FNV = 2 \measuredangle POS = 2 \phi. Hence, the remaining side of this triangle is fixed as well and given by the cosine theorem:

IV = \sqrt{NV^2 + IV^2 - 2\ NV \cdot IV \cos{\widehat{INV}}} = \sqrt{\left(\frac R 2\right)^2 + \left(\frac{R - 2r}{2}\right)...

= \sqrt{r^2 + (R^2 - 2rR)\ \frac{1 - \cos{2 \phi}}{2}} = \sqrt{r^2 + (R^2 - 2rR) \sin^2{\phi}} = \sqrt{(R - r)^2 - (R^2 - 2rR...

Since the incenter I is fixed and the distance IV is independent of the "rotation" of the triangle \triangle ABC according to Poncelet's porism, the locus of orthopoles V of the circumcircle diameter SS' is a circle (I, v) centered at the incenter I and with radius v = IV. From the above formula, it is clear that the radius of this circle is limited by r \le v \le R - r for 0 \le \phi \le 90^o.
--------------------------------------------------------
Since the angle \measuredangle FNV  = 2 \phi is fixed and the triangle \triangle INV just rotates around the incenter I, the angle \measuredangle FNV = 2 \psi is also fixed. Using the sine theorem for the triangle \triangle INV,

\frac{R}{2 \sqrt{r^2 + (R^2 - 2rR) \sin^2{\phi}}} = \frac{NV}{IV} = \frac{\sin{\widehat{NIV}}}{\sin{\widehat{INV}}} = \frac{\...

\sin{2 \psi} = \frac{R \sin{2 \phi}}{2 \sqrt{r^2 + (R^2 - 2rR) \sin^2{\phi}}}

From the above formula, it is clear that 0 \le 2 \psi < 180^o for 0 \le 2 \phi \le 180^o and that \sin{2 \psi} = 1,\ 2 \psi = 90^o for

4R^2 \sin^2{\phi}\ \cos^2{\phi} = 4\left[r^2 + (R^2 - 2rR) \sin^2{\phi}\right]

R^2 \sin^4{\phi} - 2rR\ \sin^2{\phi} + r^2 = (R\ \sin^2{\phi} - r)^2 = 0

\sin^2{\phi} = \frac r R,\ \ \cos{2 \phi} = \frac{R - 2r}{R}

The Simson lines s_P \equiv QF \perp s_P' \equiv Q'F with the poles P, P' pass through the orthopole F \in (I, r) (Feuerbach point) of the diacentral line OI \equiv PP' and they are perpendicular to each other. Likewise, the Simson lines s_S \perp s_S' with the poles S, S' pass through the orthopole V \in (I, v) of the line SS' and they are also perpendicular to each other. In addition, the Simson line pairs s_P, s_S and s_P', s_S' both form a constant angle equal to \frac \phi 2. Let V' \equiv IV \cap (I) be the intersection of the line IV with the incircle (I) and p \parallel s_S, p' \parallel s_S' 2 lines through the point V' parallel to the Simson lines s_S, s_S'. The concentric circles (I, v), (I, r) are centrally similar with homothety center I and so are the parallel lines s_S \parallel p, s_S' \parallel p' passing through the corresponding points V, V' in this homothety. It is clear that the mutually perpendicular Simson lines s_S \perp s_S' pass through fixed points of the circle (I, v) iff their mutually perpendicular parallels p \perp p' pass through fixed points of the circle (I, r) in a rotation of the central angle \measuredangle FIV \equiv \measuredangle FIV' around the incenter I by an arbitrary angle \theta. Therefore, to complete the solution of the problem, we have to prove the following lemma:
--------------------------------------------------------
Let (I) be an arbitrary circle with a fixed diameter Q, Q' and F an arbitrary point on this circle, forming a right angle triangle \triangle QFQ' with the fixed hypotenuse QQ'. Let 0 \le \measuredangle FIV = 2 \psi \le 180^o be an arbitrary central angle of this circle with a fixed measure, the point V also lying on the circle (I). Let p \perp p' be 2 perpendicular lines through the point V, both forming the same arbitrary fixed angle \frac \phi 2 \le 90^o with the perpendicular lines QF \perp Q'F. In a rotation of the point F around the circle center I by an arbitrary angle \theta, the perpendicular lines p \perp p' pass through fixed points X, X' of the circle (I), forming a right angle triangle \triangle XVX' with the fixed hypotenuse XX'.
--------------------------------------------------------
The proof of this lemma is easy: Let the perpendicular lines p \perp p' through the point V intersect the circle (I) at points X, X'. Since the angle \measuredangle XVX' = 90^o is right, XX' is a diameter of the circle (I). The triangles \triangle FIQ, \triangle VIX are both isosceles with IF = IQ = IV = IX = r, where r is radius of the circle (I). Hence, the angles \measuredangle IFQ = \measuredangle IQF are equal and the angles \measuredangle IVX = \measuredangle IXV are also equal. Let the lines FQ, VX intersect at a point Z. From the concave quadrilateral IFZV, sum of the angles \measuredangle IFQ + \measuredangle IVX is equal to

\measuredangle IFQ + \measuredangle IVX = 360^o - (\measuredangle FZV + 360^o - \measuredangle FIV) = \measuredangle FIV -\me...

From the convex quadrilateral QIXZ, the angle \measuredangle QIX is equal to

\measuredangle QIX = 360^o - (\measuredangle QZX + 180^o - \measuredangle IQF + 180^o - \measuredangle IXV) =

= \measuredangle IQF + \measuredangle IXV - \measuredangle QZX = 2 \psi - \frac \phi 2 - \frac \phi 2 = 2 \psi - \phi

Since the measures of the angles 2 \psi, \frac \phi 2 are both fixed and the point Q is also fixed, the point X is necessarily fixed as well. Since XX' is a diameter of the circle (I), both the points X, X' are fixed and they do not depend on the arbitrary angle of rotation \theta.
--------------------------------------------------------

This concludes the proof of the problem proposition. Wow !!! Cool And look mom, no cubics !

Related problem: Instead of "rotating" the triangle \triangle ABC around he fixed circumcircle (O) and incircle (I) according to Poncelet's porism, we can keep the triangle fixed and rotate the diametrally opposite poles S, S' \in (O) around the circumcenter. When the poles S, S' are identical with the intersections P, P' of the diacentral line OI with the circumcircle, i.e., \phi = \measuredangle POS = 0, the Simson line pivot points X, X' are identical with the intersections Q, Q' of the incircle with the diacentral line OI, i.e., IQ = IQ' = r. When the diameter SS' is perpendicular to the diacentral line OI, i.e., \phi = \measuredangle POS = 90^o, the Simson line pivot points X, X' are on a line perpendicular to the diacentral line OI and IX = IX' = R - r. From this, we may conjecture that the locus of the Simson line pivot points is an ellipse centered at the incenter I, with the semiminor axis IQ = r, semimajor axis IX = R - r and the focal length (using Euler's formula for R, r, d = OI)

\sqrt{(R - r)^2 - r^2} = \sqrt{R^2 - 2rR} = d = OI

I am 100 % sure that this is true, even though I have not performed the calculation yet.

Yetti
Simson_lines_30611.GIF
 Description   
 Filesize   23.11KB
 Viewed   139 Time(s)

Simson_lines_30611.GIF

_________________
Carthage must be destroyed.
Last edited by yetti on Fri Apr 29, 2005 9:24 pm; edited 4 times in total 
PostPosted: Wed Apr 27, 2005 3:21 am  Back to top 
  ProfilePM
mecrazywong
Yang-Mills Theory
Yang-Mills Theory

Offline
Joined: 07 Jul 2004
Posts: 605
Location: Hong Kong
Hong Kong

To rate posts you must be logged in
#3
Impressive work, yetti!

PostPosted: Thu Apr 28, 2005 5:43 am  Back to top 
  ProfilePMMSNICQ
yetti
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 31 Dec 2004
Posts: 2108
Location: New York
Czech Republic

To rate posts you must be logged in
#4
mecrazywong wrote:
Impressive work, yetti!

Thank you. I was thinking about this problem on and off for days, trying different approaches. At first, I thought that some cubics would be involved, because the first thing one notices is that when "rotating" the triangle around the fixed circumcircle and incircle according to Poncelet's porism, the locus of the feet of normals from a fixed pole on the circumcircle to the triangle sides (i.e., to the tangents to the fixed incircle) is by definition a limacon with both the inner and outer loops tangent to the incircle and that the Simson line pivot point necessarily lies on this limacon. The solution really surprised me.

Yetti
_________________
Carthage must be destroyed.

PostPosted: Thu Apr 28, 2005 9:15 am  Back to top 
  ProfilePM
pestich
Poincare Conjecture
Poincare Conjecture

Offline
Joined: 03 Mar 2004
Posts: 181
Botswana

To rate posts you must be logged in
#5
Sad
Last edited by pestich on Thu May 12, 2005 4:19 pm; edited 3 times in total 
PostPosted: Mon May 02, 2005 9:12 am  Back to top 
  ProfilePM
yetti
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 31 Dec 2004
Posts: 2108
Location: New York
Czech Republic

To rate posts you must be logged in
#6
pestich wrote:
Every time I come accross one of Yetti's proofs I know I have to take extra care
to avoid the 'pencil' of potential harmful side effects on the brain (the one inside the
head, the one on the shoulders) which it definitly may cause inside an unsuspecting
fresh unspoiled mind like mine. Simply put, it is like he tries to braid the however few
brain curves I have and keep them from floating free, as for the solid formations
up there, he tries to split them up and extract crumb by crumb. And he knows pretty
well that it is exactly like this, or well within epsilon of his true intentions.

Huh??? Confused

pestich wrote:

My main complaint though is that it takes too long to upload topics infected by Yetti.

The advice for your atypical case is simple:
(a) You can get a faster computer.
(b) You can complain at the JFK airport.

Yetti
_________________
Carthage must be destroyed.

PostPosted: Mon May 02, 2005 7:46 pm  Back to top 
  ProfilePM
yetti
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 31 Dec 2004
Posts: 2108
Location: New York
Czech Republic

To rate posts you must be logged in
#7

me

yetti wrote:
Related problem: Instead of "rotating" the triangle \triangle ABC around he fixed circumcircle (O) and incircle (I) according to Poncelet's porism, we can keep the triangle fixed and rotate the diametrally opposite poles S, S' \in (O) around the circumcenter. When the poles S, S' are identical with the intersections P, P' of the diacentral line OI with the circumcircle, i.e., \phi = \measuredangle POS = 0, the Simson line pivot points X, X' are identical with the intersections Q, Q' of the incircle with the diacentral line OI, i.e., IQ = IQ' = r. When the diameter SS' is perpendicular to the diacentral line OI, i.e., \phi = \measuredangle POS = 90^o, the Simson line pivot points X, X' are on a line perpendicular to the diacentral line OI and IX = IX' = R - r. From this, we may conjecture that the locus of the Simson line pivot points is an ellipse centered at the incenter I, with the semiminor axis IQ = r, semimajor axis IX = R - r and the focal length (using Euler's formula for R, r, d = OI)

\sqrt{(R - r)^2 - r^2} = \sqrt{R^2 - 2rR} = d = OI


And here is the proof, by a special request:

Both the orthopole V of the diameter SS' and the pivot point X of the Simson lines with the fixed pole S lie on the circle (I, v = IV) centered at the fixed incenter I of the variable triangle \triangle ABC. In addition, the angle \theta = \angle QIX = 2 \psi - \phi, where Q is the intersection of the diacentral line OI with the incircle (I) closer to the fixed circumcenter O.

v = IV = \sqrt{r^2 + (R^2 - 2rR) \sin^2{\phi}} = \sqrt{r^2 + d^2 \sin^2{\phi}}

\sin{2\psi} = \frac{R \sin{2 \phi}}{2v} = \frac{R \sin \phi \cos \phi}{v}

\cos{2\psi} = \sqrt{1 - \sin^2{2\psi}} = \frac{\sqrt{v^2 - R^2 \sin^2{\phi} \cos^2{\phi}}}{v} =

= \frac{\sqrt{r^2 + (R^2 - 2rR) \sin^2{\phi} - R^2 \sin^2{\phi} (1 - \sin^2{\phi})}}{v} =

= \frac{\sqrt{r^2 -2rR \sin^2{\phi} + R^2 \sin^4{\phi}}}{v} = \frac{r - R \sin^2{\phi}}{v}

We have to show that the point X lies on an ellipse with center I, semimajor axis R- r and semiminor axis r coincident with the diacentral line OI, i.e.,

(?)\ \ \frac{v^2 \cos^2{\theta}}{r^2} + \frac{v^2 \sin^2{\theta}}{(R - r)^2} = 1

regardless of the angle \theta = 2\psi - \phi. Using the trigonometric formula for \sin{(2\psi - \phi)} and the relation d = OI = R^2 - 2rR,

\frac{v^2 \cos^2{\theta}}{r^2} + \frac{v^2 \sin ^2{\theta}}{(R - r)^2} = \frac{v^2}{r^2} - \frac{v^2 \sin^2{\theta}}{r^2} + \...

= \frac{v^2}{r^2} - \frac{d^2 \left[ R^2 \sin^2{\phi} \cos^4{\phi} - 2R(r - R \sin^2{\phi}) \sin^2{\phi} \cos^2{\phi} + (r - ...

= \frac{v^2}{r^2} - \frac{d^2 \sin^2{\phi}}{r^2(R - r)^2}\ \left[R^2 (1 - 2 \sin^2{\phi} + \sin^4{\phi}) - 2rR(1 - \sin^2{\ph...

+ 2R^2 (\sin^2{\phi} - \sin^4{\phi}) + r^2 - 2rR \sin^2{\phi} + R^2 \sin^4{\phi} \left]\ =

= \frac{v^2}{r^2} - \frac{d^2 \sin^2{\phi} (R^2 - 2rR + r^2)}{r^2(R - r)^2} = \frac{v^2 - d^2 \sin^2{\phi}}{r^2} = \frac{r^2}...
_________________
Carthage must be destroyed.

PostPosted: Mon May 02, 2005 8:13 pm  Back to top 
  ProfilePM
yetti
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 31 Dec 2004
Posts: 2108
Location: New York
Czech Republic

To rate posts you must be logged in
#8
pestich wrote:
Let me say a couple of words related to the problem of Simson lines in Poncelet set-up.

It is known that the angle between Simsons of two points P1 and P2 on circumcircle is
1/2 of the central angle P1-O-P2. Now, if we have a third point P3 on the circumcircle
then the triangle XYZ made by Simsons is similar to P1P2P3 and independent of the
reference triangle ABC.

Right. For example, it was the simplest step in the above proof. But I would not venture to say "it is known" without putting forward some argument, for I did not see it in geometry textbooks dealing with Simson lines.

pestich wrote:

Poncelet configuration just provides an anker - the fixed
point I- incenter of ABC.

Huh ?? Confused
There are a lot of other fixed points, besides the incenter: the circumcenter, the entire incircle and circumcircle, the entire diacentral line OI including its intersections with the fixed incircle and circumcircle and of course, the fixed Simson line poles. What is so special about the incenter to be an anchor? Do you know any theorems about geometric anchors?

pestich wrote:

As triangle ABC rotates around incircle a la Poncelet, for the XYZ
to have fixed angles it's vertices have to move along some 3 concurrent circles
(concurrency is for the case when the 3 Simsons are concurrent) and the
sides of XYZ have to pass thru the 3 points of pairwise circles intersections.
These 3 points correspond to the points P1,P2,P3. So all Simsons for point Pi pass
thru it's circle intersection.

How do you know that the vertices of the triangle \triangle XYZ move on circles, besides finding their locus experimentally? Rest assured that the Sketchpad does it in no time and I looked for the loci of all points of interest. It is as if I said: The Simson lines of diametrally opposite points are perpendicular to each other and intersect at their orthopole. Therefore, the orthopole moves on a fixed circle and the 2 Simson lines intersect this circle at a fixed diameter. It is certainly possible, but until you show that it is also necessary, it is just possible. You would have to find either the fixed center and fixed radius of one of these circles or the common circle of similarity of the rotation centers of the spiral similarities of all triangles \triangle XYZ. The second possibility would show that the 3 circles are concurrent. You did neither.

pestich wrote:

The triangle of centers of these 3 circles is also similar to P1P2P3.

When P1P2P3 slides along the circumcircle, the 3 circles
move along so that their centers have circular orbits around the incenter of ABC.
According to Lissajus, the 3 circle intersections points are on an ellipse centered at
incenter of ABC.

You did not even show that you have 3 circles, much less that the Simson lines pass through their their intersections. And now you come up with another totally unproven claim - their centers, which you did not find at all, move on a circle concentric with the incircle. If these circles had constant radii (which they do not have), their intersections would move on (different) circles. To move on an ellipse, these circles have to - well - pulsate and on top of that, pulsate in the right way. Nothing you have said makes it possible to use the parametric equation of a closed Lissajous' curve, not to mention the amplitudes being different, the frequency ratio equal to one and the phase shift non-zero.

pestich wrote:

It seems that every time we look at a problem of 2 intersecting circles, with lines
thru the intersection point, there is a 3rd circle concurrent to the pair to complete
the picture to the fine configuration of Simson-Poncelet.

Maj. Pestich

In conclusion, you did not prove a single iota of your claims. After you have seen a proof, which is indisputable. An idea is not enough, it may or it may not result in a proof. In this case, until you find the fixed intersections of the incircle with the diacentral line to be on the Simson lines with the poles identical with the endpoints of the diacentral diameter, it is like trying to nail a piece of slime to the wall. And to find them, you need not only the properties of Simson lines, but also the special case of the 2nd Fontene theorem, leading to the identity of the orthopole of the diacentral line with the Feuerbach point, and the homotheties between the circumcircle, incircle and the 9-point circle. If you commit this kind of fallacies elsewhere, I may choose to ignore them, but if you commit them in replies to me, you can expect a refutation.

Have fun.
_________________
Carthage must be destroyed.

PostPosted: Thu May 05, 2005 1:13 pm  Back to top 
  ProfilePM
Display posts from previous:   Sort by:   
8 Posts • Page 1 of 1
Post new topic   Reply to topic View previous topicView next topic
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
You cannot post calendar events in this forum


© Copyright 2008 AoPS Incorporated. All Rights Reserved. • FoundationPrivacyContact Us