A Sallen-Key 3-Pole Butterworth Active Highpass Filter - Design Sheet (DS2)

by John-Paul Bedinger

 

      Of the various topologies you can select for making active filters, the Sallen-Key uses the least number of filter components. Furthermore, a 3-pole response (18 db/oct) is possible using only 1 op-amp. Below is a brief mathematical description on how to compute the component values for a Butterworth (steepest response with no ripple) 3-pole highpass filter with selectable output gain.

 

 Diagram 1: A 3-pole Sallen-Key highpass filter with output gain.

 

Looking at node 3b (where v3 = v3a = v3b), the node voltage equation can be re-written to be:

 

   where          and M defines the AC gain of the circuit in the passband (Eq. 1 and 2).

 

The rest of the node voltage equations can be written:

 

                                                    (Eq. 3)

            (Eq. 4)

                            (Eq. 5)

 

When solved for the filter transfer function H(s) = Vout/Vin, we get:

 

            (Eq. 6)

 

Note that the standard form of a 3-pole Butterworth highpass filter at cutoff frequency 1 rad/sec is: 

 

             (Eq.7)  

 

where  Kac is the AC gain of the filter, the same as our M.

 

Equating like terms in Equations 7 and 6 gives the solve block:

 

                                                                     (Eq.8) 

                                                     (Eq.9) 

 

      (Eq.10) 

 

               (Eq.11) 

 

We choose our output gain: Kac = M = 4 (12 dB)

 

Also, we choose values for components:  C1= 100 uF, C2= 100uF, C3= 100uF, R5= 10 k-ohms

 

Solving now for R1, R2, R3,and R4 gives (exact):

R1= 5 k-ohms, R2= 20 k-ohms, R3= 10 k-ohms, R4= 30 k-ohms

 

Rounding R1-R4 to standard EIA 1% tolerance decade values gives:

R1= 4.99 k-ohms, R2= 20 k-ohms, R3= 10 k-ohms, R4= 30 k-ohms

 

Diagram 2: A 3-pole Sallen-Key Butterworth highpass filter with cutoff at 1 rad/sec and a gain of 4 (12dB).

 


Practical Notes:

·        For different gain values you can refer to my table below, or resolve the solve block with a different value of Kac. Use solutions only with all real positive roots. If you have Mathcad(TM), you can download this MathCAD worksheet to help you.

·        C1, C2, and C3 from the table can be scaled together by a factor y, which should be done so that the source impedance Rin is much less than the magnitude of the impedance (Rin + C1) at the cutoff frequency Fc. This will set the cutoff frequency to 1/y rad/sec.

·        R1, R2, and R3 from the table can be scaled together by a factor x, which will set the cutoff to 1/(x *y) rad/sec, or:

                Fc = 1/(2*3.1416*x*y) Hz        (Eq. 12)

 

M

0dB

6dB

12dB

18dB

24dB

30dB

36dB

R1(ohms)

7180.57

5862.23

5000

4223

3531.2

2933.18

2425.61

R2(ohms)

2819.43

11533.14

20000

31821.84

49433.83

76312.84

117808

R3(ohms)

49394.66

14790.74

10000

7441.38

5728.67

4467.49

3499.48

R4(ohms)

0

10000

30000

70000

15000

31000

63000

R5(ohms)

infinite

10000

10000

10000

1000

1000

1000

C1(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

C2(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

C3(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

 

M

42dB

48dB

54dB

60dB

66dB

72dB

78dB

R1(ohms)

1998.55

1640.85

1342.31

1094.1

888.6151

719.3089

580.4752

R2(ohms)

182295.4

282964.2

440633.6

688213.9

1077780

1691800

2660920

R3(ohms)

2744.79

2153.77

1690.71

1328.07

1044.13

821.7427

647.4178

R4(ohms)

127000

25500

51100

102300

20470

40950

81910

R5(ohms)

1000

100

100

100

10

10

10

C1(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

C2(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

C3(Farads)

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

100.E-6

 Table 1: Prototype component values for a Butterworth highpass filter response at 1 rad/sec.

 


Example:

 

We want Fc= 80 Hz, C1= 0.1uF, C2= 0.1uF, and C3= 0.1uF, and a gain of 30 dB. The source resistance is 220 ohms or less.

 

Use Table 1 for 30dB prototype values, then scale y for the correct capacitor range:

 

The scale factor y is 0.1uF/100uF, or = 0.001. Thus, 80 Hz = 1/(2*3.1416*x*0.001). Solving for x gives: x = 1.989

 

Scaling R1, R2, R3, and R4 by x gives:

R1= 5834.1 ohms, R2= 151786 ohms, R3= 8885.8 ohms, R4= 31 k-ohms, R5=1 k-ohm

 

Rounding R1-R4 to standard EIA 1% tolerance decade values gives:

R1= 5.9 k-ohms, R2= 150 k-ohms, R3= 8.87 k-ohms, R4= 30.9 k-ohms, R5=1 k-ohm

 

The magnitude of (C1 + Rin) at 80Hz is:

 (1/(2*3.14*80*0.1e-6)^2+220^2)^.5 = 19.9 k-ohm at 80 Hz.

 

Since the combined impedance of 19.9 k-ohm at the cutoff frequency is much greater than the 220 ohm source resistance by itself, the value for C1 should work well.

 


Change Log:

v.1.0.2 Removed some legal rambling at the bottom of this change log – that’s all in the About This Site and General Disclaimer now. Fixed some formatting and links.

v.1.0.1 Made MathCad file a zip file.

v.1.0 Initial release.