2
\$\begingroup\$

I am stuck trying to determine the phase response of a high pass filter. I was able to find the transfer function for the high pass filter and the magnitude but I am stuck finding the phase. I found the transfer function of a high pass filter as $$\frac{V_{out}(j\omega)}{V_{in}(j\omega)}=\frac{j\omega}{j\omega+\frac{1}{RC}}$$

and I calculated the magnitude of the high pass filter as

$$|V_{out}(j\omega)|=\frac{\omega}{\sqrt{\omega^{2}+(\frac{1}{RC})^{2}}}$$

I found online that the phase response of a high pass filter is $$\frac{\pi}{2}-tan^{-1}(\omega RC)$$ But i cannot figure out how the derivation got there. I would really like to understand their logic. Thank you for anyone who can teach me the rest of this derivation

\$\endgroup\$

1 Answer 1

2
\$\begingroup\$

The phase response is just the argument of the transfer function (just as the magnitude response is the absolute value).

The argument of a quotient is the argument of the numerator minus the argument of the denominator, i.e.,

$$ \phi = \operatorname{arg} \frac{j\omega}{j\omega+\frac{1}{RC}} = \operatorname{arg}(j\omega) - \operatorname{arg}(j\omega+\frac{1}{RC}) = \frac{\pi}{2} - \operatorname{atan2}(\omega, 1/RC ) $$ which is essentially just what you wrote.

Note that I did not write \$\operatorname{tan}^{-1}(y/x)\$ but \$\operatorname{atan2(y,x)}\$ because the former is only correct if \$x\$ is positive. It is incorrect when \$x\$ is zero or negative. In your case here, when \$\omega\$ is always positive that makes no difference, but I think it is good practice to use atan2.

\$\endgroup\$
4
  • \$\begingroup\$ Thank you for responding. How does $$arg(j\omega)$$ become $$\frac{\pi}{2}$$ and $$arg(j\omega+\frac{1}{RC})$$ become $$atan2(\omega,(1/RC))$$? I have never seen this notation before on this \$\endgroup\$ Commented Sep 5, 2013 at 3:04
  • \$\begingroup\$ Well, the argument is just the angle of the complex number in the complex plane. So \$jw\$ with positive \$w\$ has argument 90 degrees or \$\pi/2\$. For complex numbers with both real and imaginary part (\$x + jy\$) this is given by the arcustangent \$tan^{-1}(y/x)\$, when x is positive. In the general case one has to make slight adjustments and hence the atan2 function. The name is because it is named like this in many programming languages (see eg en.wikipedia.org/wiki/Atan2). See also en.wikipedia.org/wiki/Arg_%28mathematics%29 for general information. \$\endgroup\$
    – Andreas H.
    Commented Sep 5, 2013 at 3:13
  • \$\begingroup\$ Ahhh yes this is slowly coming back to me now. I remember learning this in my system dynamics and controls class as an undergrad but that was 3 years ago :/ I am a mechanical engineer so it's good to see the electrical engineering people help me out! \$\endgroup\$ Commented Sep 5, 2013 at 3:17
  • \$\begingroup\$ sure, no problem \$\endgroup\$
    – Andreas H.
    Commented Sep 5, 2013 at 3:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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