# Parametric equation of a circle intersecting 3 points

Tags:
1. Nov 14, 2016

### whitejac

1. The problem statement, all variables and given/known data
Given the points P0 = (0,a), P1 = (b,0), P2 = (0,0), write the parametric equation of a circle that intersects the 3 points.

Assume that b > a and both are positive.

2. Relevant equations
X = h + rcos(t)
Y = k + rsin (t)
r = √((x-h)2 + (y-k)2
Cos (t) = (x-h)/r
Sin (t) = (y-k) / r

3. The attempt at a solution
I'm having a really hard time visualizing parametric equations. I drew the circle and know that the challenge is finding the center, but I don't know how to express it. the circle is shifted right and up. I don't know how to express the coordinates of the center in terms of P0 and P1 as well as so it would intersect both these two points as well as (0,0).

2. Nov 14, 2016

### Krylov

The three given points $P_0$, $P_1$ and $P_2$ are known. They all have the same unknown distance $r > 0$ to the unknown center $(h,k)$. Can you use this to set up and solve a system of equations in these unknowns?

3. Nov 14, 2016

### BvU

A useful property of a circle is that the center point is the same distance from all points on the rim. So where are the points that have the same distance from (0,0) and (b,0) ? same for (0,0) and (0,a) ?

4. Nov 14, 2016

### whitejac

Would it look something like this..?

At the origin:
R = √(h2 + k2)

At point (b,0):
R =√(b-h)2 + k2)

At point (0,a):
R = √(a-k)2 + h2)

5. Nov 14, 2016

### BvU

Yes. Not just 'something like' : that's what it is, exactly (*)
Now solve.

(*) assuming case insensitivity -- or did you have a purpose changing from r to R ?

6. Nov 14, 2016

### whitejac

So...

k = √(r2 - h2)

implies

r2 = (b-h)2 + r2 - h2

Which reduces to:

h = b2 / 2

If you go back and find

h = √(r2 - k2)

then you'd find

k = a2 / 2

7. Nov 15, 2016

### BvU

Must be mistaken. One way to see that is from dimensions (this a physics community, after all ): left and right have different dimensions (length and length2, respectively).

Back to the drawing board...

8. Nov 15, 2016

### Pratik89

It was incorrect, those coordinates belong to the centroid

Last edited: Nov 15, 2016
9. Nov 15, 2016

### LCKurtz

No, that is not where the circumcenter is. The points form a right triangle so the hypotenuse is a diameter. That is what makes the problem easier.

10. Nov 15, 2016

### Pratik89

11. Nov 16, 2016

### ehild

Making a sketch, one sees that it is really a right triangle and where the center of the circle is, and to find the radius is also very easy.

12. Nov 16, 2016

### BvU

Pratik wants answers instead of questions, not guidance to help him find it by himself (and thus gain some insight). OK.
Well, the points that are equidistant from (0,0) and (b,0) are on x=b/2.
Once you know what you are looking for, a parametric equation isn't so hard to write down.
Erszebeths's picture is worth a thousand words ... or more.

And the culture at PF should remain as it is in this respect: learn a person to fish instead of giving him/her a fish.