Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

How can I generate all triangles which have integral sides and area, and exactly two of its three sides are equal?

For example, a triangle with sides ${5,5,6}$ satisfies these terms.

share|improve this question

2 Answers 2

up vote 3 down vote accepted

Heron's formula says the area of a triangle with sides $a, b, c$ is $$Area = \sqrt{s(s-a)(s-b)(s-c)}$$ where $s$ is the semiperimeter, $s = \frac{a+b+c}{2}$. Now, your assumption is two sides are equal, so $a, b, b$. The area is now $$Area = \sqrt{s(s-a)(s-b)^2} = (s-b)\sqrt{s(s-a)}$$ Now $s-b$ might not be an integer, if $s$ is not, but it will at worst be a fraction of the form $\frac{z}{2}$. I will deal more with this later.

For now, we want $\sqrt{s(s-a)}$ to be an integer, so $s(s-a)$ must be a perfect square. Now, $s = \frac{a + 2b}{2} = \frac{a}{2} + b$ so this simplifies to wanting $$s(s-a) = (\frac{a}{2} + b)(\frac{a}{2} + b - a) = (b + \frac{a}{2})(b - \frac{a}{2}) = b^2 - \frac{a^2}{4}$$ a perfect square, so let's say it equals $c^2$. Thus, we want all integer solutions to $$c^2 + \left(\frac{a}{2}\right)^2 = b^2$$ If we happen to get a solution such that $(s-b)\sqrt{s(s-a)}$ is not an integer, then multiply all side lengths by 2 to get a similar triangle with $s$ and thus $s-b$ an integer.

Other than that, the problem is reduced to finding all solutions to this equation, which is a well known problem with a well known solution.

share|improve this answer
    
Does $ s(s-a)$ being a perfect square, $\implies s-b$ also integer? –  lab bhattacharjee Sep 17 '12 at 15:30
    
@labbhattacharjee Thanks! fixed –  Graphth Sep 17 '12 at 15:50

Area of isosceles triangle with sides $a,a,b$ is $\frac{b\sqrt{4a^2-b^2}}{4}$

Now $\sqrt{4a^2-b^2}$ must be integer$=c$(say),

$\implies 4a^2-b^2=c^2\implies b^2+c^2=4a^2$

Clearly, $b,c$ are of same parity.

If $b,c$ are odd $=2C+1,2D+1$ respectively, then $b^2+c^2=(2C+1)^2+(2D+1)^2$ $=4(C^2+D^2+C+D)+2\equiv 2\pmod 4$,not multiple of $4$.

$\implies b,c$ are even

Let $b=2B,c=2C$

So, $B^2+C^2=a^2$

Area of the isosceles triangle becomes $B\sqrt{a^2-B^2}=BC$

The parametric solution of $B^2+C^2=a^2$ is $k(p^2-q^2), 2pqk, k(p^2+q^2)$ where $p,q,k$ are any non-negative integers and $p>q$ .

So, the general solutions are $a=k(p^2+q^2), b=2B=2k(p^2-q^2),c=2C=2(2pqk)=4pqk$ or $a=k(p^2+q^2), b=2B=4pqk,c=2C=2k(p^2-q^2)$, the area being $pq(p^2-q^2)$, in either case.

If $k=1, p=2,q=1,a=2^2+1^2=5,b=2(2^2-1^2)=6$ or $b=4\cdot 2\cdot 1=8$

share|improve this answer
    
just double a pythagorean right triangle –  i. m. soloveichik Sep 17 '12 at 15:05
    
@i.m.soloveichik, exactly, the connecting edge being any one of the sides other than hypotenuse. –  lab bhattacharjee Sep 17 '12 at 16:12
1  
@i.m.soloveichik But, what if the triangle you double has area that is, say $9/2$, or say a side length that is irrational. If that's all you did, you would miss these. –  Graphth Sep 17 '12 at 16:32
    
@Graphth, the general solution will always produce integral sides and area as $p,q,k$ are integers, right? –  lab bhattacharjee Sep 19 '12 at 13:08

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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