Phasor Representation of Sinusoidal Variables

In the following discussion about AC circuit analysis, all sinusoidal variables (currents and voltages) are assumed to be of the same frequency. In general, arithmetic operations of sinusoidal functions are not convenient as they will involve using trigonometric identities. For example, given two sinusoids of the same frequency $\omega$:

$\displaystyle \left\{ \begin{array}{l} x(t)=A\cos(\omega t+\phi)\\
y(t)=B\cos(\omega t+\psi) \end{array} \right.$ (16)

we can find their sum as:
$\displaystyle z(t)=x(t)+y(t)$ $\displaystyle =$ $\displaystyle A\cos(\omega t+\phi)+B\cos(\omega t+\psi)$  
  $\displaystyle =$ $\displaystyle A\cos\omega t\cos\phi-A\sin\omega t\sin\phi
+B\cos\omega t\cos\psi-B\sin\omega t\sin\psi$  
  $\displaystyle =$ $\displaystyle (A\cos\phi+B\cos\psi) \cos\omega t
-(A\sin\phi+B\sin\psi)\sin\omega t$ (17)

If we further assume

$\displaystyle \left\{\begin{array}{l}
A\cos\phi+B\cos\psi=C\cos\theta=C_r\\
A\sin\phi+B\sin\psi=C\sin\theta=C_i
\end{array}\right.$ (18)

then $z(t)$ above can be written as the following in terms of $C$ and $\theta$:

$\displaystyle z(t)=C\cos\theta \cos\omega t+C\sin\theta \sin\omega t
=C \cos(\omega t+\theta)$ (19)

To find $C$ and $\theta$ we note that

$\displaystyle \left\{\begin{array}{l}
C_r^2+C_i^2=C^2\\
\\
C_i/C_r=\tan\theta
\end{array}\right.$    

and solve these equations to get

$\displaystyle C=\sqrt{C_r^2+C_i^2},\;\;\;\;\;\;\;\;\;\;\;
\theta=\tan^{-1} (C_i/C_r)$ (20)

Alternatively, the addition of the two sinusoidal functions above can also be carried out when they are treated as the real (or imaginary) parts of rotating vectors in the complex plane, and thereby more conveniently added in vector forms in the complex plane:

$\displaystyle z(t)=x(t)+y(t)$ $\displaystyle =$ $\displaystyle A\cos(\omega t+\phi)+B\cos(\omega t+\psi)$  
  $\displaystyle =$ $\displaystyle Re(A e^{j(\omega t+\phi)})+Re(B e^{j(\omega t+\psi)})$  
  $\displaystyle =$ $\displaystyle Re[(A e^{j\phi}+B e^{j\psi})e^{j\omega t}]$ (21)

We can get $z(t)$ in three steps:

Here the two complex exponentials $A e^{j\phi}$ and $B e^{j\psi}$ are called the phasors of the sinusoidal functions $x(t)$ and $y(t)$.

RotatingVectors.gif

phasors.png

Specifically, the sum $z(t)=x(t)+y(t)$ of the two sinusoidal functions once represented in phasor form in complex plane can be found as the real part of the vector sum in the following three steps:

(review of complex arithmetic)

PhasorBlockDiagram1.png

Example

Consider three sinusoidal voltage sources $v_1(t)=6\;\sin(\omega t)$, $v_2(t)=12\;\sin(\omega t+\pi/2)$ and $v_3(t)=4\;\sin(\omega t-\pi/2)$ in series. According to the KVL, the overall voltage will be the algebraic sum of the three:

$\displaystyle v(t)=v_1(t)+v_2(t)+v_3(t)
=6\;\sin(\omega t)+12\;\sin(\omega t+\pi/2)
+4\;\sin(\omega t-\pi/2)$ (29)

sinusoidalsum.gif

While the addition of these sinusoidal functions is not easy to carry out (still remember all the trigonometric identities?), it is quite straight forward to find the vector sum if the voltages are represented as phasors:

$\displaystyle \dot{V}$ $\displaystyle =$ $\displaystyle \dot{V}_1+\dot{V}_2+\dot{V}_3
=6\angle 0^\circ+12\angle 90^\circ+4\angle -90^\circ$  
  $\displaystyle =$ $\displaystyle 6+j8=10 \angle \left[\tan^{-1}(8/6)\right]
=10\angle\; 53.1^\circ$ (30)

The resulting voltage is $v(t)=10\;sin(\omega t+53.1^\circ)$

vectorsum.gif

sinusoidalsum1.gif

Phasor and the Fourier transform

A sinusoidal function $x(t)=A\cos(\omega t+\phi)$ can be expressed in either Fourier transform (Fourier series) or phasor representation:

$\displaystyle x(t)$ $\displaystyle =$ $\displaystyle A\cos(\omega t+\phi)=Re[Ae^{j\phi}\;e^{j\omega t}]=Re[\dot{X} e^{j\omega t}]$  
  $\displaystyle =$ $\displaystyle \left(\frac{A}{2}e^{j\phi}\right)e^{j\omega t}
+\left(\frac{A}{2}e^{-j\phi}\right)e^{-j\omega t}
=X_1 e^{j\omega t}+X_{-1}e^{-j\omega t}$ (31)

where

$\displaystyle \dot{X}=Ae^{j\phi},\;\;\;\;\;\;\;\;\;\;\;\;\;\;
X_1=\frac{A}{2}e^{j\phi},\;\;\;X_{-1}=\frac{A}{2}e^{-j\phi}$ (32)

We see that the phasor $\dot{X}$ and the Fourier coefficients $X_1$ and $X_{-1}$ are essentially the same, in the sense that they are both coefficients representing the amplitude and phase of the complex exponential function $e^{j\omega t}$.

In both cases the real time function can be obtained (inverse Fourier transform) by either adding the two complex vectors $X_1e^{j\omega t}$ and $X_{-1}e^{-j\omega t}$ rotating in opposite directions (CCW and CW), or simply taking the real part of the complex variable $\dot{X}e^{j\omega t}$