I know that phasors are used to simplify the calculations, I also get why in AC, current in a capacitor leads the voltage by 90°, and lags for an inductor (we can see that by differentiating/integrating), but how do we know it will solve differential equations too? Thinking of sine and cosine terms as vectors really helps in adding/subtracting them. (But I don't see how it helps in multiplying/dividing)

Suppose I have an RL circuit connected in series to an AC emf like this:

enter image description here

Then, by phasors, we would say $\overline{Z} = \sqrt{R^2 + L^2 \omega^2}\angle\tan^{-1}(\omega L/R)$ or $\overline{i} = \displaystyle\frac{e_0}{\sqrt{R^2 + L^2 \omega^2}}\angle\tan^{-1}(-\omega L/R)$ as $\overline{i}=\displaystyle\frac{\overline{e}}{\overline{Z}}$ (Taking $\overline{e} = e_0\angle0$)

but without using phasors, we would have written $+e_0\sin(\omega t)-iR - L\frac{\text{d}i}{\text{d}t} = 0$ and try to solve that differential equation, right?

How do we know we're solving that differential equation by doing phasor algebra?

By the expression I got for $\overline{i}$, $i(t) = \displaystyle\frac{e_0}{\sqrt{R^2 + L^2 \omega^2}} \sin(\omega t - \tan^{-1}(\omega L/R))$

Then I thought, is it really the solution to the differential equation $+e_0\sin(\omega t)-iR - L\frac{\text{d}i}{\text{d}t} = 0$?

I looked up the solution to that equation and it's slightly different $$i(t) = \displaystyle\frac{e_0}{\sqrt{R^2 + L^2 \omega^2}} \sin(\omega t - \tan^{-1}(\omega L/R)) - c e^{-Rt/L}$$ According to this , the phasor answer is correct if c=0, which does not mean $i(0) = 0$, I checked the graph, but it isn't too different from the phasor answer for positive values of t. Why are we getting this extra term different from the phasor answer?


$\sin(\omega t)$ can be represented as the imaginary part of $e^{i\omega t}$, which is a rotating vector in the complex plane (called phasors). Coincidentally, the solutions of the differential equations involved in the circuit come out to be of the form $e^{i\omega t}$, and so we can take the real part of the equation and prove it.

Now suppose $g$ is a solution of $+e_0\sin(\omega t)-iR - L\frac{\text{d}i}{\text{d}t} = 0$, and $f$ is a solution of $iR +L\frac{\text{d}i}{\text{d}t} = 0$, the you can easily verify that $g+\lambda f$ is also a solution $+e_0\sin(\omega t)-iR - L\frac{\text{d}i}{\text{d}t} = 0$ because it is a linear equation. We can solve the DE for $f$ and get $f=ce^{\frac{-Rt}{L}}$

  • $\begingroup$ Ok, but the solution to the differential equation has an extra +e^-Rt/L term in it? $\endgroup$ – Rick Mar 28 '18 at 22:01
  • $\begingroup$ @Rick I've edited the answer $\endgroup$ – Prathyush Poduval Mar 29 '18 at 8:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.