3

I've been doing this problem for a good while now and just coming to grips with it. I'd like to know if my solution is acceptable. Thanks in advance.

A forced and damped harmonic oscillator satisfies $$\ddot{x}+2k\dot{x}+\omega_0^2 x = F\cos\omega t$$ where $\omega_0^2 > 2k^2.$ Determine the amplitude of the long term solution as a function of $\omega$.

The associated homogeneous equation is

$$\ddot{x}+2k\dot{x}+\omega_0^2 x =0$$

The characteristic polynomial of our homogenous equation is

$$P(\lambda)= \lambda^2 + 2k\lambda +\omega_0^2$$

and has roots

$$\lambda_{\pm}= -k\pm \sqrt{k^2-\omega_0^2} =-k\pm i\sqrt{\omega_0^2-k^2}$$

The question tells us that $\omega_0^2 > 2k^2.$ This is the under-damped case ($\omega_0^2 > k^2$). The solution to the homogeneous equation is then

$$x_h(t)=e^{-kt}(A\cos\omega' t+ B\sin\omega' t)$$

where $\omega'=\sqrt{\omega_0^2-k^2}$

Now, to find the particular solution we use the complex method.

We need to solve

$$\ddot{z}+2k\dot{z}+\omega_0^2 z = Ae^{i\omega t}$$

By using the trial function $z(t)= ae^{i\omega t}$ we get

$$\ddot{z}+2k\dot{z}+\omega_0^2 z=[(i\omega)^2+2k(i\omega)+\omega_0^2]ae^{i\omega t}=Ae^{i\omega t}$$

Let's define $P(i\omega) = [(i\omega)^2+2k(i\omega)+\omega_0^2] $ since it's our characteristic equation.

The complex solution is

$$z(t) = \dfrac{1}{P(i\omega)}Ae^{i\omega t}$$.

The question want us to find the long-term solution. This is also called the steady-state solution. This is the real solution of $z(t)$. Letting $x_p(t)$ be the real particular solution, then what we are looking for is simply

$$x_p(t) = Re(z(t))$$.

What is the real part of $z(t)$? We can find this using complex polar coordinates.

$$\begin{align} P(i\omega)& = (i\omega)^2 + 2k(i\omega)+\omega_0^2\\ & = (\omega_0^2-\omega^2) +2k\omega i \end{align}$$

Recall that $R = \sqrt{x^2 +y^2} = \sqrt{(\omega_0^2-\omega^2)^2 +4k^2\omega^2}$

Now we can rewrite $P(i\omega)$ as

$$P(i\omega) = Re^{i\theta}=R(\cos \theta + i\sin \theta)$$

Now we only want the real part, we don't want $i\sin \theta$

$$\begin{align}x_p(t) & = \dfrac{1}{Re^{i\theta}}Ae^{i\omega t} \\ & = \dfrac{1}{R}Ae^{i(\omega t - \theta)} \end{align}$$

The real solution to $z(t)$ is

$$x_p(t) = \dfrac{1}{R}A\cos(\omega t-\theta)$$

letting $G(\omega) = \dfrac{1}{R}$ we have

$$x_p(t) = G(\omega)A\cos(\omega t - \theta)$$.

the amplitude of the long-term solution as a function of $\omega$ can be written as

$$H(\omega) = G(\omega)A = \dfrac{A}{\sqrt{(\omega_0^2-\omega^2)^2 +4k^2\omega^2}}$$

where $G(\omega)A$ is the amplitude of $x_p$

Is this too long an answer? I might get something like this in the exam and I think this could take a long time to write.

What is a resonance curve and if I was expected to sketch one how would I go about it. I think I saw that it's the amplitude against $\omega$. Thanks very much.

1 Answers1

2

$$z(t) = \frac{1}{P(i \omega)}Ae^{i \omega t} = \frac{P(-i \omega)}{P(i \omega)P(-i \omega)}A(\cos({ \omega t})+i\sin(\omega t))$$ $P(-i \omega)$ is the complex conjugate of $P(i\omega) $, so the denominator is now real. So $$Re(z (t)) = \frac{1}{P(i \omega)P(-i \omega)} [Re(P(-i \omega))\cos(\omega t)+Im(P (-i \omega))\sin (\omega t)]$$ And $$A \cos (\omega t) + B \sin(\omega t) = \sqrt{A^2 + B^2} \sin(\omega t + \theta) $$ Where $$\sin(\theta) = \frac{A}{ \sqrt{A^2+B^2}} $$

infinitylord
  • 4,777