Active filters

Kevin Weddle

Feb 23, 2004
1,620
Joined
Feb 23, 2004
Messages
1,620
Some active filters are designed at resonance and some are not. It would seem that resonace would cause more problems in circuits. Doesn't a high pass filter followed by a low pass filter and an amplifier produce the same result without the added harmonics?

 

audioguru2

Apr 6, 2004
12,026
Joined
Apr 6, 2004
Messages
12,026
An active filter adds some positive feedback at the cutoff frequency so the response is flat to the cutoff then a sharp slope.

A multiple-feedback bandpass filter can have any Q you want. A simple highpass filter then a simple highpass filter is a very poor bandpass filter. Both filters attenuate the frequency you want.

View attachment 40704

 

Kevin Weddle

Feb 23, 2004
1,620
Joined
Feb 23, 2004
Messages
1,620
An active filter doesn't require positive feedback. Designing simple LR filters and RC filters in a cascade arrangement along with amplifiers might create less harmonics and varying harmonic frequencies.

 

audioguru2

Apr 6, 2004
12,026
Joined
Apr 6, 2004
Messages
12,026
An active filter doesn't require positive feedback. Designing simple LR filters and RC filters in a cascade arrangement along with amplifiers might create less harmonics and varying harmonic frequencies.
Cascaded RL or RC filters have a droopy response. An active filter adds positive feedback to make the response sharp at the cutoff frequency. Then the response is flat up to the cutoff frequency then a sharp drop past the cutoff frequency.

The multiple-feedback bandpass filter opamp circuit I posted uses positive feedback to increase the Q of the simple RC filters. You cannot have a high Q without the positive feedback.

Todays opamps have extremely low harmonics that can barely be measured.
All the harmonics together plus noise of an OPA134 opamp is 0.00008% distortion.
 

Kevin Weddle

Feb 23, 2004
1,620
Joined
Feb 23, 2004
Messages
1,620
Since an LC filter can only produce 40db/decade attenuation, it can be replaced with a two pole RC or LC filter.

A simple low pass filter will always attenuate a higher frequency. A tuned filter might attenuate a higher or lower frequency within the fundamental frequency bandwidth depending on conditions. Resonance is often avoided in circuit design, but is also chosen to reduce signal loss.

 

audioguru2

Apr 6, 2004
12,026
Joined
Apr 6, 2004
Messages
12,026
An RC low-pass filter is -3dB at the cutoff frequency and starts reducing the level at 1/5th its cutoff frequency.
Two buffered RC filters in series are -6dB at the cutoff frequency and start reducing frequencies that are 1/10th the cutoff frequency.

A Butterworth 2nd-order (two poles) active lowpass filter uses positive feedback to boost the cutoff frequency so it is down only -3dB and so that the respose is flat to about 1/4 the cutoff frequency.

 

Kevin Weddle

Feb 23, 2004
1,620
Joined
Feb 23, 2004
Messages
1,620
[quote A Butterworth 2nd-order (two poles) active lowpass filter uses positive feedback to boost the cutoff frequency so it is down only -3dB and so that the respose is flat to about 1/4 the cutoff frequency.

 

audioguru2

Apr 6, 2004
12,026
Joined
Apr 6, 2004
Messages
12,026
I was wrong.
A high-order Butterworth lowpass filter is almost perfect but a second-order Butterworth filter
attenuates signals above about half its cutoff frequency. Curves are shown in Google but they are difficult to see because they have orders fro 1 to 5 all at the same time.

 

Kevin Weddle

Feb 23, 2004
1,620
Joined
Feb 23, 2004
Messages
1,620
I built a simple parallel LC and found about only 1Kohm resistance at resonance. Is this typical for power supply size components?

 

Hero999

Oct 28, 2007
2,433
Joined
Oct 28, 2007
Messages
2,433
Providing the schematic might help us answer your question.

 
Top