Bode plots for simple transfer functions
This lecture also appears in Control: an introduction.
It turns out that bode plots, both magnitude and phase, given their logarithmic scale (recall that the \(\omega\)-axes are also plotted logarithmically), are quite asymptotic to straight-lines for first- and second-order systems. Furthermore, higher-order system transfer functions can be re-written as the product of those of first-and second-order. For instance, $$\begin{align} H(s) &= \frac{\emp[1em] s + \emp[1em]}{s^3 + \emp[1em] s^2 + \emp[1em] s + \emp[1em]} \\ &= \emp[1em] \cdot \left( \emp[1em] s + 1 \right) \cdot \frac{1}{\emp[1em] s + 1} \cdot \frac{1}{s^2 + \emp[1em] s + \emp[1em]} \end{align}$$
Recall (from, for instance, phasor representation) that for products of complex numbers, phases \(\phi_i\) add and magnitudes \(M_i\) multiply. For instance, \[\begin{aligned} M_1 \angle{\phi_1} \cdot \frac{1}{M_2 \angle{\phi_2}} \cdot \frac{1}{M_3 \angle{\phi_3}} = \frac{M_1}{M_2 M_3} \angle{\left(\phi_1-\phi_2-\phi_3\right)}. \end{aligned}\] And if one takes the logarithm of the magnitudes, they add; for instance, \[\begin{aligned} \log\frac{M_1}{M_2 M_3} = \log M_1 - \log M_2 - \log M_3. \end{aligned}\] There is only one more link in the chain: first- and second-order Bode plots depend on a handful of parameters that can be found directly from transfer functions. There is no need to compute \(\abs{H(j\omega_0)}\) and \(\angle H(j\omega_0)\)!
In a manner similar to , we construct Bode plots for several simple transfer functions in this lecture. Once we have these simple “building blocks,” we will be able to construct sketches of higher-order systems by graphical addition because logarithmic magnitudes and phases combine by summation, as shown in .
Constant gain
For a transfer function that is simply a constant real gain \(H(s) = K\), the frequency response function is trivially \(H(j\omega) = K\). Its magnitude \(\abs{H(j\omega)} = \abs{K}\). For positive gain \(K\), the phase is \(\angle{H(j\omega)} = 0\), and for negative \(K\), the phase is \(\angle{H(j\omega)} = 180 \deg\).
Pole and zero at the origin
In , we have already demonstrated how to derive from the transfer function \(H(s) = s\), a zero at the origin, the frequency response function plotted in . Similarly, for \(H(s) = 1/s\), a pole at the origin, the frequency response function plotted in figure 13.2.
Bode plots for (a) a pole at the origin and (b) a zero at the origin.
Real pole and real zero
The derivations for real poles and zeros are not included, but the resulting Bode plots are shown in figure 13.3.
Bode plots for (a) a single real pole and (b) a single real zero.
Complex conjugate pole pairs and zero pairs
The derivations for complex conjugate pole pairs and zero pairs are not included, but the resulting Bode plots are shown in figure 13.4.
Bode plots for (a) a complex conjugate pole pair and (b) a complex conjugate zero pair.
Online Resources for Section 13.4
No online resources.