How to change a Sallen-Key low pass filter so that it becomes a shelving filter?
I.e. I want it to accentuate the bass frequencies (by 10dB, transfer frequency around 150Hz), all higher frequencies should pass without amplification but no attenuation either.
I found this webpage for dimensioning (using Butterworth characteristics), it dimensions a Sallen-Key filter:
I made an attempt to build the filter in one opamp circuit. Simulations in PSpice 9.1 prove that it must work. However, I would like to have someone's feedback on it. Are there any problems expected with this kind of solution?
Here is an image of PSpice schematic + simulation (frequency response):
The high-pass part is a Sallen-Key 2nd order filter. The low-pass is a simple bass boost capacitor that I added in the negative feedback. I added a resistor to ground to make this part work. The Sallen-Key filter should still work as a 6dB/octave because in its frequency area, the bass boost filter has become flat.
Could someone please take a look and comment on it?
Thanks, U.c.
It'll work fine. It -was- a SK low pass filter but the heavy massaging you've given it in spice means it's a one off. I.e you'd have a lot of work calculating that particular response shape from scratch :)
Try Filter Pro software from TI, and it's free
martin
This circuit will work for the bass boost with the fixed response. However in most of cases the bass boost has to be made adjustable: the boost frequency, gain and Q should be set independently. In this design, all of the parameters are tightly coupled, so it wouldn't be possible.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
I think a number must have been transcribed wrong. The circuit as posted looks to me to be in oscillation.
Read my book.
Hi MooseFET.
I think it appears so because of the peak, however, the filter consists of 2 parts really. It won't oscillate because the Sallen-Key filter is safely dimensioned as a Butterworth curve. The lower part around the 390n accounts for the bass boost.
The filter is meant to operate at this fixed setting, btw.
Unless I misread the values in the picture, it certainly isn't Butterworth. The gain of the opamp is just about 4 at the cut off and the resistor ratio is about 1:2.
For the unity gain case, the resistor ratio would be correct for a Butterworth but not for the case of a gain near 4.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required