next up previous
Next: The Bode plot of Up: active_filter Previous: Source and load

Second order systems

Consider the following circuits composed of R, C, L connected in series:

rcl_2nd_order.gif

By Kirchhoff's Voltage law, we have

\begin{displaymath}v_{in}(t)=V_R + v_L + v_C = L \; \frac{di(t)}{dt} + R \; i(t) + \frac{1}{C} \int_0^t i(\tau) d\tau \end{displaymath}

where $v_{in}(t)$ is the input voltage across all three elements. Taking a time derivative of the equation, we get:

\begin{displaymath}L \; \frac{d^2i(t)}{dt^2} + R \; \frac{di(t)}{dt} + \frac{1}{C} i(t)
=\frac{dv_{in}(t)}{dt} \end{displaymath}

If the current is treated as the output $y(t)=i(t)$, this LTI system's response to a complex exponential input $v_{in}(t)=e^{st}$ is $i(t)=H(s)e^{st}$, and the equation becomes:

\begin{displaymath}L \frac{d^2}{dt^2} e^{st}H(s)+R \frac{d}{dt} e^{st}H(s) +\frac{1}{C} e^{st}H(s)=\frac{d}{dt} e^{st} \end{displaymath}

i.e.,

\begin{displaymath}H(s)\;(s L e^{st}+R e^{st} +\frac{1}{sC} e^{st})=e^{st} \end{displaymath}

The transfer function can be found to be

\begin{displaymath}H(s)=\frac{I(s)}{V_{in}(s)}=\frac{1}{sL+R+(sC)^{-1}} \end{displaymath}

Alternatively, the impedances of R, C and L can be expressed in Laplace domain as:

\begin{displaymath}Z_R=R,\;\;\;Z_C=1/sC,\;\;\;\;Z_L=sL \end{displaymath}

and the total impedance of the RCL series circuit is

\begin{displaymath}Z_{total}(s)=Z_R(s)+Z_C(s)+Z_L(s)=R+\frac{1}{sC}+sL \end{displaymath}

we have

\begin{displaymath}V_{in}(s)=Z_{total}(s) I(s)=[Z_L(s)+Z_R(s)+Z_C(s)]I(s)
=(sL+R+\frac{1}{sC})I(s) \end{displaymath}

and we get the same transfer function:

\begin{displaymath}H(s)=\frac{I(s)}{V_{in}(s)}=\frac{1}{sL+R+(sC)^{-1}} \end{displaymath}

If we consider the voltage across one of the three element (R,C,L) as the output, the circuit can be considered as a voltage divider with transfer function

\begin{displaymath}H(s)=\frac{Z_{out}(s)}{Z_{total}(s)} \end{displaymath}

where $Z_{out}(s)$ depends on how the output is defined. If the output is the voltage across L,

\begin{displaymath}H_L(s)=\frac{V_L(s)}{V_{in}(s)}=\frac{sL}{sL+R+(sC)^{-1}}
=\frac{s^2}{s^2+2\zeta \omega_n s + \omega_n^2} \end{displaymath}

If the output is the voltage across R,

\begin{displaymath}H_R(s)=\frac{V_R(s)}{V_{in}(s)}=\frac{R}{sL+R+(sC)^{-1}}
=\frac{2\zeta \omega_n s}{s^2+2\zeta \omega_n s + \omega_n^2} \end{displaymath}

If the output is the voltage across C,

\begin{displaymath}H_C(s)=\frac{V_C(s)}{V_{in}(s)}=\frac{(sC)^{-1}}{sL+R+(sC)^{-1}}
=\frac{\omega_n^2}{s^2+2\zeta \omega_n s + \omega_n^2} \end{displaymath}

where

\begin{displaymath}\left\{ \begin{array}{ll}
\omega_n \stackrel{\triangle}{=} ...
...{L}} &
\mbox{is the damping coefficient}
\end{array} \right.
\end{displaymath}

In general, the transfer function of a second order system can be written as

\begin{displaymath}H(s)=\frac{N(s)}{D(s)}=\frac{N(s)}{s^2+2\zeta \omega_n s + \omega_n^2}
=\frac{C_1}{s-s_1}+\frac{C_2}{s-s_2} \end{displaymath}

where $s_1$ and $s_2$ are the two roots of the denominator $D(s)$:

\begin{displaymath}s_{1,2}=(-\zeta \pm j \sqrt{1-\zeta^2}) \omega_n = -\zeta \omega_n \pm j\omega_d \end{displaymath}

with

\begin{displaymath}\omega_d \stackrel{\triangle}{=}\omega_n \sqrt{1-\zeta^2} < \omega_n
\end{displaymath}

is called the damped natural frequency. The coefficients $C_1$ and $C_2$ can be found by partial fraction extension.


next up previous
Next: The Bode plot of Up: active_filter Previous: Source and load
Ruye Wang 2008-11-24