1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Loci of centres in Ellipse

  1. Aug 14, 2015 #1
    20150814085536.jpg Dear Friends!
    If we have a variable triangle formed by joining a point of ellipse and its two focii shown in figure,how to find locii of various centres.Can you please suggest some reference to find these .I have tried by using simple method for incentre locus which I left as it goes long. Though I am still thinking.
  2. jcsd
  3. Aug 14, 2015 #2


    Staff: Mentor

    First off, locus is singular and loci (one 'i') is plural, not "locii."

    For your ellipse, the two foci are at ##(\pm c, 0)##, where ##c = \sqrt{a^2 - b^2}##). Possibly you could write an equation for the locus of the various centers as the point P on the ellipse moves around.
  4. Aug 15, 2015 #3


    User Avatar
    Science Advisor

    The center of such a triangle is the point whose coordinates are the means of the corresponding coordinates of the vertices.

    Here, you have a ellipse given by [itex]\frac{x^2}{a^2}+ \frac{y^2}{b^2}= 1[/itex]. Assuming that a> b, the two foci have coordinates [itex](-\sqrt{a^2- b^2}, 0)[/itex] and [itex](\sqrt{a^2- b^2}, 0)[/itex]. You can write a point on that ellipse as [itex]P= (a cos(t), b sin(t)[/itex] for parameter t between 0 and [itex]2\pi[/itex]. "Average" those coordinates.

    (That turns out to be remarkably simple and gives a very easy answer!)
    Last edited by a moderator: Aug 15, 2015
  5. Aug 16, 2015 #4
    Thanks Mark44 and Hallsoflvy!
  6. Aug 19, 2015 #5

    Attached Files:

  7. Aug 19, 2015 #6
    Dear friends ! In the solution of finding locus of incentre ,I am getting parabola as locus which does not seem to be by physical picturization.
  8. Aug 19, 2015 #7


    User Avatar
    Science Advisor

    Well, what did you do when you followed our suggestions? What is the "average" of the three points [itex](-\sqrt{a^2- b^2}, 0)[/itex], [itex](\sqrt{a^2- b^2}, 0)[/itex], and [itex](a cos(t), b sin(t))[/itex]? It is NOT a parabola! Please show exactly what you did.
  9. Aug 19, 2015 #8
    Thanks Hallsfloy! I believe you are telling to find centroid locus which I am showing now. It is an ellipse which is fine.The first locus which I had calculated was of incentre in attached picture earliar.

    Attached Files:

  10. Aug 19, 2015 #9


    User Avatar
    Science Advisor

    Yes, that's right. As I said originally, "That turns out to be remarkably simple and gives a very easy answer!" The y coordinates of the first two points are 0 and their x coordinates cancel so the equation of the locus is just the original ellipse, divided by 3. It is the ellipse with major and minor axes 1/3 the length of the original ellipse.
  11. Aug 19, 2015 #10
    Thanks! Now my doubt in #6 is still left!
  12. Aug 19, 2015 #11


    User Avatar
    Science Advisor

    What "doubt" do you mean? The difficulty you stated in #6 was that you got a parabola which, of course, can't be right because a parabola would have to go outside the original ellipse. But in #8 you say you got an ellipse as the answer.
  13. Aug 19, 2015 #12
    Incenter is the intersection point of bisector lines of the angles of the triangle
    Circumcenter is the intersection point of the sides perpendicular bisectors
    Ortocenter is the intersection point of the height
    Centroid is the intersection point of the medians

    The average of the coordinates gives the centroid.
    The other points are less trivial to find
    Incenter can be found using the formula here http://www.mathopenref.com/coordincenter.html

    Circumcenter P has coordinates
    x_P=\frac{{x_A}^2 {x_B}-{x_A}^2 {x_C}-{x_A} {x_B}^2+{x_A} {x_C}^2-{x_A} {y_B}^2+{x_A} {y_C}^2+{x_B}^2 {x_C}-{x_B} {x_C}^2+{x_B} {y_A}^2-{x_B} {y_C}^2-{x_C} {y_A}^2+{x_C} {y_B}^2}{2 (-{x_A} {y_B}+{x_A} {y_C}+{x_B} {y_A}-{x_B} {y_C}-{x_C} {y_A}+{x_C} {y_B})}


    \frac{{x_A}^2 ({x_C}-{x_B})+{x_A} \left({x_B}^2-{x_C}^2+{y_B}^2-{y_C}^2\right)-{x_B}^2 {x_C}+{x_B} \left({x_C}^2-{y_A}^2+{y_C}^2\right)+{x_C} ({y_A}-{y_B}) ({y_A}+{y_B})}{2 ({x_A} ({y_B}-{y_C})+{x_B} ({y_C}-{y_A})+{x_C} ({y_A}-{y_B}))}

    Orthocenter O has coordinates

    x_O=\frac{{x_A} ({x_B} ({y_B}-{y_A})+{x_C} ({y_A}-{y_C}))-({y_B}-{y_C}) ({x_B} {x_C}+({y_A}-{y_B}) ({y_A}-{y_C}))}{{x_A} ({y_B}-{y_C})+{x_B} ({y_C}-{y_A})+{x_C} ({y_A}-{y_B})}

    y_O=\frac{{y_C} (-{x_A} {y_A}+{x_B} {y_B}+{x_C} ({y_A}-{y_B}))+({x_A}-{x_B}) (({x_A}-{x_C}) ({x_B}-{x_C})+{y_A} {y_B})}{{x_A} ({y_B}-{y_C})+{x_B} ({y_C}-{y_A})+{x_C} ({y_A}-{y_B})}

    It can be quite hard to work it out...

    Graphically it can be seen as in the following picture
    Last edited: Aug 19, 2015
  14. Aug 19, 2015 #13
    I am referring to my first picture where I have evaluated locus of incentre.
  15. Aug 19, 2015 #14
    Picture in #5
  16. Aug 19, 2015 #15
    Thanks Raffaelel! I have calculated incentre locus in attached image in #5.Need to get it verified.
    Last edited: Aug 19, 2015
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Threads - Loci centres Ellipse Date
B Ellipse tangent line using projections Nov 19, 2017
I Centre of Gravity offset Apr 17, 2017
I General formula for centre of mass of polygons Jul 15, 2016
Coordinate transformation - NED and ECEF frames Oct 6, 2015
Find centre of circle Oct 6, 2011