This filter is composed of four sections labeled “a”, “b”, “c”, and “d”. The schematic for each section is shown in the rectangular frame. The output of the first (a) section is connected to the input of the second (b), the second to the third, and the third to the fourth. Although the schematic of each section is the same, the component values aren’t. The resistors and capacitors in the first section are referred to as Ra, Rga, and Ca, the second section designations are Rb, Rgb, and Cb, the third Rc, Rgc, and Cc, and the fourth Rd, Rgd, and Cd.
Entering a value for the cutoff frequency (-3db or 0.707 point) will find the optimum resistor and capacitor values for all sections. By default the solver finds values for a Bessel implementation. For general information about filters and specific details about filter implementations, see the Filter Types article. You can change the implementation by entering a value for Type. The following table lists the acceptable values for Type and the filter implementation for each Type value.
Unless otherwise specified, the capacitor’s tolerance is assumed to be 5%. If you wish to specify a different tolerance, enter it as a percentage (10% = 10).
Even though solutions are provided for a number tolerance bands, this filter design (Sallen-Key or VCVS) performs best when the component values are more closely matched. The use of 1% tolerance resistors and reasonably high grade film capacitors are recommended. In addition, it is best to use resistor values that are greater than 10K (other than Rg) to minimize frequency errors.
If you are interested in the details, an excellent downloadable resource is available an Application Report from Texas Instruments. Search for SLOA049B, “Active Low-Pass Filter Design” by Jim Karki. Excellent, and concise, coverage of the subject matter is also given in the Art of Electronics by Horowitz and Hill.