Impedance and Generalized Ohm's Law

Impedance of Basic Components

In time domain, the relationship between the sinusoidal current through and the sinusoidal voltage across a capacitor or an inductor is described by a differential equation. However, in frequency domain, where these sinusoidal variables are represented as complex exponentials, and the components such as R, C, and L are all represented by their impedances, then the relationship between the sinusoidal voltage and current can be described by an algebraic equation.

Specifically, we represent the sinusoidal voltage and current as the projection of the corresponding vector in the complex plane rotating counter-clock wise onto the real axis:

$\displaystyle v(t)=V_p\cos(\omega t+\phi)=Re[ V_p e^{j(\omega t+\phi)} ],
\;\;\;\;\;\;\;\;
i(t)=I_p\cos(\omega t+\psi)=Re[ I_p e^{j(\omega t+\psi)} ]$ (33)

and define the impedance of a component (R, C, and L) as the ratio of the voltage and current associated with the component both in complex exponential form:

Impedance$\displaystyle =Z=\frac{V_pe^{j(\omega t+\phi)}}{I_pe^{j(\omega t+\psi)}}
=\frac...
...ac{\dot{V}}{\dot{I}}
=\frac{\mbox{Phasor of voltage}}{\mbox{Phasor of current}}$ (34)

This is the generalized version of the Ohm's law for AC circuits.

ImpedanceFT.png

One way to remember the phase between the voltage $E$ and current $I$ associated with capacitor $C$ and inductor $L$ is “ELI the ICE man”. Also, consider two extreme cases:

In a DC circuit, each resistor is measured by either its resistance $R$ or its conductance $G=1/R$. In an AC circuit each component (capacitor, inductor, or resistor) is measured by its impedance $Z$, of which the real and imaginary parts are respectively the resistance $R$ and reactance $X$, or its admittance $Y=1/Z$, of which the real and imaginary parts are respectively the conductance $G$ and susceptance $B$, as summarized below:

$\displaystyle \left\{\begin{array}{rcl}
\mbox{Impedance} &=& \mbox{Resistance} ...
...x{Susceptance} \\
\mbox{Admittance}&=&1/\mbox{Impedance}\\
\end{array}\right.$ (50)

Impedance $Z$ and admittance $Y=1/Z$ are both complex variables. The real parts $Re[Z]=R$ and $Re[Y]=G$ are always positive, but the imaginary parts $Im[Z]=X$ and $Im[Y]=B$ can be either positive or negative. Therefore $Z$ and $Y$ can only be in the 1st or the 4th quadrants of the complex plane.

In particular, the admittances of the three types of elements R, L and C are

$\displaystyle Y_R=\frac{1}{R},\;\;\;\;
Y_L=\frac{1}{Z_L}=\frac{1}{j\omega L}=\frac{-j}{\omega L},\;\;\;\;
Y_C=\frac{1}{Z_C}=\frac{1}{1/j\omega C}=j\omega C$ (51)

Ohm's law can also be expressed in terms of admittance as well as impedance. Sometimes it is more convenient in circuit analysis to use admittance instead of impedance.

Generalized Ohm's law and Kirchhoff's Laws

In general, all methods such as Ohm's law and Kirchhoff's Laws used for DC circuits composed of resistors can be generalized to AC circuits composed of capacitors, inductors, as well as resistors, all represented by their impedances. Also, if we assume all voltages and currents in a circuit are sinusoids of same frequency $\omega$, they can be represented as complex phasors.

The Ohm's law can be generalized to become:

$\displaystyle Z=\frac{\dot{V}}{\dot{I}},\;\;\;\;\;\;\dot{I}=\frac{\dot{V}}{Z},
\;\;\;\;\;\dot{V}=Z\dot{I}$ (54)

and the Kirchhoff's laws can now be stated as:

Solving AC circuit by phasor method

If only the steady state solutions of the DE describing an AC circuit is of interest, the phasor method can be used to solve the problem algebraically without solving the DEs. Specifically, all sinusoidal variables are represented as phasors in terms of their amplitudes and phases, and all components in the circuit (L and C, as well as R) are represented by their impedances, so that all the laws (Ohm's law, KCL and KVL, current and voltage dividers, parallel and series combinations of components) and methods (loop current and node voltage methods, Thevenin's and Norton's theorems, etc.) discussed for DC circuit can be applied.

phasordiagram.gif

Operations on sinusoidal variables based on the trigonometric identities are in general lengthy and tedious. The phasor method can convert such sinusoidal variables to vectors in complex plane and thereby simplify the operations.

Here is a review of complex arithmetic.

Example 1:

Solve the circuit below. The voltage from the generator is $v(t)=28.3\;\cos(5000t+45^\circ)\;V=20\sqrt{2}\;\cos(5000t+45^\circ)$.

phasorexample.gif

The given voltage $v(t)\;V$ can be expressed in phasor form as

$\displaystyle \dot{V}=28.3 e^{j\,45 \circ}
=28.3\;\angle{45^\circ}=20\sqrt{2}\;\angle{45^\circ}
=20+j\;20$ (55)

First find the impedances and admittances of the components and the two branches. As $\omega=5000$, we get

Next find all currents in phasor form: Verify:

VectorPlot.png

Example 2:

A current $i(t)=17\;\cos(1000t+90^\circ)=12\sqrt{2}\cos(1000t+90^\circ)$ flows through a circuit composed of a resistor $R=18\Omega$, a capacitor $C=83.3\mu F=83.3\times 10^{-6}F$, and an inductor $L=30 mH=30\times
10^{-3}H$ connected in series. Find the resulting voltage across all three elements.

Alternatively, we can find voltage across each of the elements:
$\displaystyle \dot{V}_R$ $\displaystyle =$ $\displaystyle \dot{I}Z_R=216\sqrt{2}\;j=305.5\;\angle{90^\circ}$  
$\displaystyle \dot{V}_C$ $\displaystyle =$ $\displaystyle \dot{I}Z_C=144\sqrt{2}\;\angle{0^\circ}=203.6$  
$\displaystyle \dot{V}_L$ $\displaystyle =$ $\displaystyle \dot{I}Z_L=360\sqrt{2}\;\angle{180^\circ}=510\;\angle{180^\circ}$ (60)

or in time domain:
$\displaystyle v_R(t)$ $\displaystyle =$ $\displaystyle 305.5\; \cos(1000t+\pi/2)$  
$\displaystyle v_C(t)$ $\displaystyle =$ $\displaystyle 203.6\; \cos(1000t)$  
$\displaystyle v_L(t)$ $\displaystyle =$ $\displaystyle 509\; \cos(1000t+\pi)$ (61)

Adding $\dot{V}_R$, $\dot{V}_C$, and $\dot{V}_L$, we get the total voltage which is the same as what we got above:

$\displaystyle \dot{V}_{total}=\dot{V}_R+\dot{V}_C+\dot{V}_L
=216\sqrt{2}\angle{90^\circ}+216\sqrt{2}\angle{180^\circ}=432 \angle{135^\circ}$ (62)

Example 3:

In the circuit below, $v_S(t)=A\cos(\omega t)=A\cos(10^6 t)$ with some unknown peak value $A$, $R_1=R_2=R_3=R=10\Omega$, and $L=10\,\mu H$. The RMS value of $v_2$ across $R_2$ is 10 V. It is also known that $v_s(t)$ and $i_s(t)$ are in phase.

Problems9.png

Solution

Problems8.png

We first note that $\dot{V}_C$ is behind $\dot{V}_2=\dot{I}_2R_2$ by $\pi/2$, and $\dot{V}_L$ is ahead of $\dot{V}_3=\dot{I}_3R_3$ by $\pi/2$ (“ELI the ICE man”). Also, as $v_s(t)$ and $i_s(t)$ are in phase, the parallel combination of the RL and RC branches introduces no phase shift, i.e., its impedance shown below must be real:

$\displaystyle Z_{RL}(\omega)\vert\vert Z_{RC}(\omega)=\frac{(R+jX_L)(R+jX_C)}{R+jX_L+R+jX_C}
=\frac{R^2-X_LX_C+j(X_L+X_C)}{2R+j(X_L+X_C)}$ (63)

i.e., $X_C+X_L=0$. We therefore get $1/j\omega C=-j\omega L=-j 10$ and $C=10^{-7}=0.1\mu F$, $Z_{RL}=10+j\,10$, $Z_{RC}=10-j\,10$, $\vert Z_{RL}\vert=\vert Z_{RC}\vert=\sqrt{10^2+10^2}=10\sqrt{2}$, i.e.,

$\displaystyle Z_{RL}(\omega)\vert\vert Z_{RC}(\omega)=\frac{(10+j10)(10-j10)}{10+j10+10-j10}=10$ (64)

As $\vert R_2\vert=\vert Z_C\vert=10 \Omega$, $\vert\dot{V}_C\vert=\vert\dot{V}_R\vert=10V$. But as they are $\pi/2$ apart in phase, we have $\vert\dot{V}_{cd}\vert=\vert\dot{V}_{RC}\vert=\vert\dot{V}_{RL}\vert=10\sqrt{2}$, and from the vector diagram $\vert\dot{V}_{ab}\vert=10\sqrt{2}$. We also get the currents through RC and RL branches are:

$\displaystyle \vert\dot{I}_{RC}\vert=\frac{\vert\dot{V}_{RC}\vert}{\vert Z_{RC}...
...t{I}_{RL}\vert=\frac{\vert\dot{V}_{RL}\vert}{\vert Z_{RL}\vert}=\frac{10}{10}=1$ (65)

But their phase difference is $\pi/2$, we have

$\displaystyle \vert\dot{I}_s\vert=\vert\dot{I}_{RL}+\dot{I}_{RC}\vert=\sqrt{2}$ (66)

The voltage across $R_1$ is $V_1=RI_s=10\sqrt{2}$, and

$\displaystyle \dot{V}_s=\dot{V}_1+\dot{V}_{cd}=10\sqrt{2}+10\sqrt{2}=20\sqrt{2}$ (67)

The peak value is therefore $A=\sqrt{2} V_s=40\,V$