0
$\begingroup$

I'm not very experienced with double limits and now I have to evaluate the limits $$ \lim_{x \to \infty,\ y \to \infty} \left[ \tanh(x+2y) + \tanh(x-2y) \right] $$ $$ \lim_{x \to \infty,\ y \to \infty} \left[ \tanh(x+2y) - \tanh(x-2y) \right] $$ and I'm really not sure how to proceed with this. Naturally the first terms are $$ \lim_{x \to \infty,\ y \to \infty} \tanh(x+2y) = 1 $$ but that's as far as I have gotten.

Does this limit exist?

EDIT: So after a little work I can show that $$ \tanh(x-2y) = \frac{\sinh(x-2y)}{\cosh(x-2y)} = \frac{e^{x-2y} - e^{-(x-2y)}}{e^{x-2y} + e^{-(x-2y)}} = \frac{e^{-2y} - e^{-2x}e^{2y}}{e^{-2y} + e^{-2x} e^{2y}} $$ which for constant $y < \infty$ behaves as $$ \lim_{x \to \infty} \tanh(x-2y) = \frac{e^{-2y} - e^{-2x}e^{2y}}{e^{-2y} + e^{-2x} e^{2y}} = \frac{e^{-2y} - 0 \cdot e^{2y}}{e^{-2y} + 0 \cdot e^{-y}} = 1 $$ but also $$ \lim_{y \to \infty} \tanh(x-2y) = \frac{e^{x-4y} - e^{-x}}{e^{x-4y} + e^{-x}} = \frac{0 - e^{-x}}{0 + e^{-x}} = - 1 \mathrm{.} $$

However as $y \to \infty$ AND $x \to \infty$ there's the term $e^{-2x}e^{2y}$ where the two terms tend to zero and infinity. What is the specific reason that allows me to say that this limit doesn't exist?

$\endgroup$
  • $\begingroup$ Is it $+\infty $ or $-\infty $ $\endgroup$ – hamam_Abdallah May 26 '17 at 15:13
  • $\begingroup$ It is $+ \infty$ $\endgroup$ – 655321 May 26 '17 at 15:15
1
$\begingroup$

Notice that

$$\begin{align}\tanh(x+2y)+\tanh(x-2y)&=\frac{\sinh(x+2y)}{\cosh(x+2y)}+\frac{\sinh(x-2y)}{\cosh(x-2y)}\\&=\frac{\sinh(x+2y)\cosh(x-2y)+\sinh(x-2y)\cosh(x+2y)}{\cosh(x+2y)\cosh(x-2y)}\\&=\frac{\sinh(2x)}{\cosh(x+2y)\cosh(x-2y)}\\&=\frac{\frac{e^{2x}-e^{-2x}}2}{\frac{(e^{x+2y}+e^{-(x+2y)})(e^{x-2y}+e^{-(x-2y)})}4}\\&=2\times\frac{1-e^{-6x}}{(e^{2y}+e^{-2x-2y})(e^{-2y}+e^{-2x-2y})}\end{align}$$

Then note that as $x\to\infty$ and $y$ is held constant, then it goes to $2$. Likewise, as $y\to\infty$ and $x$ is held constant, then it goes to $0$. Thus, the limit doesn't exist.

$\endgroup$
  • $\begingroup$ But if $y$ is held constant ($ < \infty $) isn't $\lim_{x \to \infty} \left[ \tanh(x+2y) + \tanh(x-2y) \right] = 1 + 1 = 2$? And not infinity? $\endgroup$ – 655321 May 26 '17 at 15:28
  • $\begingroup$ The denominator quickly goes to 0 in that case. $\endgroup$ – Ryan A May 26 '17 at 16:00
  • $\begingroup$ Yes but $\sinh(x+2y) \cosh(x-2y) + \sinh(x-2y) \cosh(x+2y) = \sinh(2x) \neq \sinh(4x)$ and the last term in the denominator should be $e^{-(2x-2y)} = e^{-2x+2y}$ so $$\tanh(x+2y) - \tanh(x-2y) = 2 \cdot \frac{1-e^{-4x}}{(e^{2y} + e^{-2x-2y})(e^{-2y}+e^{-2x+2y})} \mathrm{.}$$ I find it hard to believe that a sum on $\tanh$-functions would tend to infinity as $\tanh(x) \in (-1, 1)$. Anyway I think that this limit really doesn't exist. $\endgroup$ – 655321 May 26 '17 at 17:08
  • $\begingroup$ @655321 Thanks for pointing out my error. And yes, the limit doesn't exist anyways. $\endgroup$ – Simply Beautiful Art May 26 '17 at 22:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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