TABLE OF CONTENTS Steve Winder
| Note | I don't use a real high-pass filter every year, but I do use some audio coupling capacitors, and I do use sample-and-difference circuits in my testing. And that's a form of high-pass filter, so we will explain that. Because, as Steve Winder points out here, it's important to use a good "poly" cap, either polystyrene or polypropylene, not polyester./rap |
This chapter describes how to design an analog active or passive highpass filter with almost any desired specification. Examples for most types of highpass filter are given. Formulae will be presented for the denormalization of component values given in previously presented tables.
Passive highpass filters are designed using the normalized lowpass model. The model is normalized for a passband that extends from DC to 1 rad/s and is terminated with a 1 ? load resistance. The first part of the process is to carry out the conversion to a highpass model; this can then be scaled for the desired load impedance and cutoff frequency. The highpass model has a passband that extends from 1 rad/s to infinity (in theory, at least). In practice, parasitic components exist to reduce the upper frequency response. These parasitic components are, for example, capacitance between wires in an inductor's windings or inductance in the leads of a capacitor.
Converting the lowpass model into a highpass equivalent is not too demanding in all-pole filters, like Butterworth or Chebyshev types. The process requires replacing each inductor in the lowpass...
