MHB Find all possible values of |x+y+z|

  • Thread starter Thread starter anemone
  • Start date Start date
anemone
Gold Member
MHB
POTW Director
Messages
3,851
Reaction score
115
Let $x,\,y,\,z$ be complex numbers such that $|x|=|y|=|z|=1$.

If $\dfrac{x^2}{yz}+\dfrac{y^2}{zx}+\dfrac{z^2}{xy}=1$, find the product of all possible values of $|x+y+z|$.
 
Mathematics news on Phys.org
anemone said:
Let $x,\,y,\,z$ be complex numbers such that $|x|=|y|=|z|=1$.

If $\dfrac{x^2}{yz}+\dfrac{y^2}{zx}+\dfrac{z^2}{xy}=1$, find the product of all possible values of $|x+y+z|$.
[sp]Write the equation as $x^3 + y^3 + z^3 = xyz$. If $\lambda$ is any complex number with $|\lambda|=1$ then the equation is unaltered if we replace $x,y,z$ by $\lambda x,\lambda y,\lambda z$, and $|x+y+z|$ is then also unaltered. So choose $\lambda$ to be a cube root of $(xyz)^{-1}$. The equation then becomes $x^3 + y^3 + z^3 = xyz = 1.$

Now let $a=x^3$, $b=y^3$, $c=z^3$. Then $|a| = |b| = |c| = 1$ and $a+b+c = abc = 1$, from which $\frac12(a+b) = \frac12(1-c).$ So the points $1,a,b,c$ all lie on the unit circle, and the midpoint of the chord from $a$ to $b$ is the same as the midpoint of the chord from $1$ to $-c$. But the midpoint of a chord of a circle uniquely determines its endpoints. So $a$ and $b$ must be the same as $1$ and $-c$. Therefore the three points are $\{1,c,-c\}.$ But their product is $1$, so that $-c^2 = 1$, $c = \pm i.$

Therefore $x,y,z$ are cube roots of $1,i,-i$. That leaves only finitely many possible choices for $x,y,z$. You can check that all of them are obtained (by rotation of the circle) from two basic solutions $\{1,\omega,\omega^{-1}\}$ and $\{1,\omega^5,\omega^{-5}\}$, where $\omega = e^{i\pi/6}$. The corresponding values of $|x+y+z|$ are $\sqrt3+1$ and $\sqrt3-1$, with product $(\sqrt3+1)( \sqrt3-1) = \boxed2.$[/sp]
 
Opalg said:
[sp]Write the equation as $x^3 + y^3 + z^3 = xyz$. If $\lambda$ is any complex number with $|\lambda|=1$ then the equation is unaltered if we replace $x,y,z$ by $\lambda x,\lambda y,\lambda z$, and $|x+y+z|$ is then also unaltered. So choose $\lambda$ to be a cube root of $(xyz)^{-1}$. The equation then becomes $x^3 + y^3 + z^3 = xyz = 1.$

Now let $a=x^3$, $b=y^3$, $c=z^3$. Then $|a| = |b| = |c| = 1$ and $a+b+c = abc = 1$, from which $\frac12(a+b) = \frac12(1-c).$ So the points $1,a,b,c$ all lie on the unit circle, and the midpoint of the chord from $a$ to $b$ is the same as the midpoint of the chord from $1$ to $-c$. But the midpoint of a chord of a circle uniquely determines its endpoints. So $a$ and $b$ must be the same as $1$ and $-c$. Therefore the three points are $\{1,c,-c\}.$ But their product is $1$, so that $-c^2 = 1$, $c = \pm i.$

Therefore $x,y,z$ are cube roots of $1,i,-i$. That leaves only finitely many possible choices for $x,y,z$. You can check that all of them are obtained (by rotation of the circle) from two basic solutions $\{1,\omega,\omega^{-1}\}$ and $\{1,\omega^5,\omega^{-5}\}$, where $\omega = e^{i\pi/6}$. The corresponding values of $|x+y+z|$ are $\sqrt3+1$ and $\sqrt3-1$, with product $(\sqrt3+1)( \sqrt3-1) = \boxed2.$[/sp]

What a neat solution, bravo, Opalg! And thanks for participating.

I would also like to share the solution that comes along with this good problem, here goes:

Let $s=x+y+z$

Then

$\begin{align*}s^3&=x^3+y^3+z^3+3(x^2y+xy^2+y^2z+yz^2+z^2x+zx^2)+6xyz\\&=xyz\left(\dfrac{x^2}{yz}+\dfrac{y^2}{zx}+\dfrac{z^2}{xy}+3\left(\dfrac{y}{x}+\dfrac{y}{z}+\dfrac{z}{y}+\dfrac{z}{x}+\dfrac{x}{z}\right)+6\right)\\&=xyz\left(1+\left(3(x+y+z)\left(\dfrac{1}{x}+\dfrac{1}{y}+\dfrac{1}{z}\right)-9\right)+6\right)\\&=xyz\left(3s\left(\dfrac{1}{x}+\dfrac{1}{y}+\dfrac{1}{z}\right)-2\right)\\&=xyz(3s\overline{s}-2)\,\,\text{because $\overline{s}=\overline{x}+\overline{y}+\overline{z}=\dfrac{1}{x}+\dfrac{1}{y}+\dfrac{1}{z}$}\\&=xyz(3\left| s \right|^2-2)\end{align*}$

Taking absolute values, we find $\left| s \right|^3=\left| \left| 3s \right|^2-2 \right|$. It follows that $\left| s \right|$ must be a positive real root of $a^3-3a^2+2=0$ or $a^3+3a^2-2=0$. However, since the negative real roots of $a^3-3a^2+2=0$ are exactly the additive inverses of the positive real roots of $a^3-3a^2+2=0$, and all three roots of $a^3-2a^2+2=0$ are real ($a^3-3a^2+2=0$ maybe factored as $(a-1)(a^2-2a-2)=0$, and the discriminant of $a^2-2a-2$ is positive), the product of all possible values of $\left| s \right|$ is $(-2)\cdot(-1)^n$, where $n$ denotes the number of negative real roos of $a^3-3a^2+2=0$. By Descartes's Rule of Signs, we see that $n$ is odd, so the answer is 2, as desired.
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
I'm interested to know whether the equation $$1 = 2 - \frac{1}{2 - \frac{1}{2 - \cdots}}$$ is true or not. It can be shown easily that if the continued fraction converges, it cannot converge to anything else than 1. It seems that if the continued fraction converges, the convergence is very slow. The apparent slowness of the convergence makes it difficult to estimate the presence of true convergence numerically. At the moment I don't know whether this converges or not.
Back
Top