Phase Relations in Active Filters Frequency in radians per second is equal to 2π times frequency

in Hz ( f ), since there are 2π radians in a 360° cycle. Because the

expression is a dimensionless ratio, either f or ω could be used.
By Hank Zumbahlen [[email protected]]
The center frequency can also be referred to as the cutoff
In applications that use filters, the amplitude response is frequency (the frequency at which the amplitude response of
generally of greater interest than the phase response. But in some the single-pole, low-pass filter is down by 3 dB—about 30%).
applications, the phase response of the filter is important. An In terms of phase, the center frequency will be at the point at
example of this might be where a filter is an element of a process which the phase shift is 50% of its ultimate value of –90° (in this
control loop. Here the total phase shift is of concern, since it may case). Figure 2, a semi-log plot, evaluates Equation 1 from two
affect loop stability. Whether the topology used to build the filter decades below to two decades above the center frequency. The
produces a sign inversion at some frequencies can be important. center frequency (=1) has a phase shift of –45°.
It might be useful to visualize the active filter as two cascaded 0 180
filters. One is the ideal filter, embodying the transfer equation;
the other is the amplifier used to build the filter. This is illustrated
in Figure 1. An amplifier used in a closed negative-feedback loop

PHASE ANGLE (INVERTED)

PHASE ANGLE (IN-PHASE)
can be considered as a simple low-pass filter with a first-order
response. The gain rolls off with frequency above a certain
breakpoint. In addition, there will be, in effect, an additional
180° phase shift at all frequencies if the amplifier is used in the –45 135
inverting configuration.
FILTER AMPLIFIER
TRANSFER FUNCTION TRANSFER FUNCTION
Ff (s) Fa (s)

Figure 1. Filter as cascade of two transfer functions.

–90 90
Filter design is a two-step process. First, the filter response is 0.01 0.1 1 10 100
chosen; then, a circuit topology is selected to implement it. The NORMALIZED FREQUENCY
filter response refers to the shape of the attenuation curve. Often,
this is one of the classical responses such as Butterworth, Bessel, Figure 2. Phase response of a single-pole, low-pass filter
or some form of Chebyshev. Although these response curves are about the center frequency (in-phase response, left axis;
usually chosen to affect the amplitude response, they will also inverted response, right axis).
affect the shape of the phase response. For the purpose of the
Similarly, the phase response of a single-pole, high-pass filter is
comparisons in this discussion, the amplitude response will be
given by:
ignored and considered essentially constant.
Filter complexity is typically defined by the filter “order,” which is (2)
related to the number of energy storage elements (inductors and

capacitors). The order of the filter transfer function’s denominator
defines the attenuation rate as frequency increases. The asymptotic Figure 3 evaluates Equation 2 from two decades below to two
filter rolloff rate is –6n dB/octave or –20n dB/decade, where n decades above the center frequency. The normalized center
is the number of poles. An octave is a doubling or halving of the frequency (=1) has a phase shift of +45°.
frequency; a decade is a tenfold increase or decrease of frequency. So
a first-order (or single-pole) filter has a rolloff rate of –6 dB/octave It is evident that the high-pass and the low-pass phase responses
or –20 dB/decade. Similarly, a second-order (or 2-pole) filter has a are similar, only shifted by 90° (π/2 radians).
rolloff rate of –12 dB/octave or –40 dB/decade. Higher-order filters 90 –90
are usually built up of cascaded first- and second-order blocks. It is,
of course, possible to build third- and, even, fourth-order sections
with a single active stage, but sensitivities to component values PHASE ANGLE (INVERTED)
PHASE ANGLE (IN-PHASE)

and the effects of interactions among the components on the

frequency response increase dramatically, making these choices
less attractive.
45 –135
The Transfer Equation
First, we will take a look at the phase response of the transfer
equations. The phase shift of the transfer function will be the
same for all filter options of the same order.
For the single-pole, low-pass case, the transfer function has a
phase shift, ϕ, given by
0 –180
0.01 0.1 1 10 100
(1) NORMALIZED FREQUENCY

Figure 3. Phase response of a single-pole, high-pass filter
where: ω = frequency (radians per second) with a center frequency of 1 (in-phase response, left axis;
ω0 = center frequency (radians per second) inverted response, right axis).

http://www.analog.com/analogdialogue
Analog Dialogue 41-10, October (2007) 
For the second-order, low-pass case, the transfer function has a Again, it is evident that the high-pass and low-pass phase responses
phase shift that can be approximated by are similar, just shifted by 180° (π radians).
In higher-order filters, the phase response of each additional
(3) section is cumulative, adding to the total. This will be discussed in
greater detail later. In keeping with common practice, the displayed
where α is the damping ratio of the filter. It will determine the phase shift is limited to the range of ±180°. For example, –181° is
peaking in the amplitude response and the sharpness of the really the same as +179°, 360° is the same as 0°, and so on.
phase transition. It is the inverse of the Q of the circuit, which
First-Order Filter Sections
also determines the steepness of the amplitude rolloff or phase
First-order sections can be built in a variety of ways. The most
shift. The Butterworth has an α of 1.414 (Q of 0.707), producing
straightforward way is illustrated in Figure 6, simply using a
a maximally flat response. Lower values of α will cause peaking
passive R-C configuration. The center frequency of this filter
in the amplitude response.
is 1/(2πRC). It is commonly followed by a noninverting buffer
0 180 amplifier to prevent loading by the circuit following the filter,
which could alter the filter response. In addition, the buffer can
provide some drive capability. The phase will vary with frequency
–45 135 as shown in Figure 2, with 45° phase shift at the center frequency,

PHASE ANGLE (INVERTED)

PHASE ANGLE (IN-PHASE)

exactly as predicted by the transfer equation, since there are no

extra components to modify the phase shift. That response will be
–90 90
referred to as the in-phase, first-order, low-pass response. The buffer
will add no phase shift, as long as its bandwidth is significantly
greater than that of the filter.
–135 45
R
IN OUT
+
C

–180 0
0.01 0.1 1 10 100
NORMALIZED FREQUENCY
Figure 6. Passive, low-pass filter.

Figure 4. Phase response of a 2-pole, low-pass filter with Remember that the frequency in these plots is normalized, i.e.,
a center frequency of 1 (in-phase response, left axis; the ratio to the center frequency. If, for example, the center
inverted response, right axis). frequency were 5 kHz, the plot would provide the phase response
to frequencies from 50 Hz to 500 kHz.
Figure 4 evaluates this equation (using α = 1.414) from two decades An alternative structure is shown in Figure 7. This circuit,
below to two decades above the center frequency. Here the center which adds resistance in parallel to continuously discharge an
frequency (=1) shows a phase shift of –90°. integrating capacitor, is basically a lossy integrator. The center
The phase response of a 2-pole, high-pass f ilter can be frequency is again 1/(2πRC). Because the amplifier is used in
approximated by the inverting mode, the inversion introduces an additional
180° of phase shift. The input-to-output phase variation with
frequency, including the amplifier’s phase inversion, is shown
(4) in Figure 2 (right axis). This response will be referred to as the

inverted, first-order, low-pass response.
In Figure 5 this equation is evaluated (again using α = 1.414), from
C
two decades below to two decades above the center frequency (=1), R
+

IN OUT
which shows a phase shift of –90°. R
180 0

135 –45
PHASE ANGLE (INVERTED)
PHASE ANGLE (IN-PHASE)

Figure 7. Active, single-pole, low-pass filter using

an op amp in the inverting mode.
90 –90
T he c i r c u it s s how n above, wh ic h at te nu ate t he h i g h
frequencies and pass the low frequencies, are low-pass filters.
Similar circuits also exist to pass high frequencies. The passive
45 –135 configuration for a first-order, high-pass filter is shown in
Figure 8; and its phase variation with normalized frequency
is shown in Figure 3 (in-phase response).
0 –180
C
0.01 0.1 1 10 100
+

IN OUT
NORMALIZED FREQUENCY
R
Figure 5. Phase response of a 2-pole, high-pass filter
with a center frequency of 1 (in-phase response, left axis;
inverted response, right axis). Figure 8. Passive, high-pass filter.

 Analog Dialogue 41-10, October (2007)

The plot in Figure 3 (left axis) will be referred to as the in phase, The amplifier gain in Sallen-Key filters can be increased by
first-order, high-pass response. The active configuration of the connecting a resistive attenuator in the feedback path to the
high-pass filter is shown in Figure 9. The phase variation with inverting input of the op amp. However, changing the gain will
frequency is shown in Figure 3 (right axis). This will be referred affect the equations for the frequency-determining network,
to as the inverted, first-order, high-pass response. and the component values will have to be recalculated. Also, the
amplifier’s dynamics are more likely to need scrutiny, since they

+
IN OUT
C R R introduce gain into the loop.

The Multiple-Feedback (MFB), Low-Pass Filter

The multiple feedback filter is a single-amplifier configuration
Figure 9. Active, single-pole, high-pass filter. based on an op amp as an integrator (an inverting configuration)
inside a feedback loop (see Figure 12). Therefore, the dependence
Second-Order Sections of the transfer function on the op amp parameters is greater than
A variety of circuit topologies exists for building second-order in the Sallen-Key realization. It is hard to generate high-Q,
sections. To be discussed here are the Sallen-Key, the multiple- high-frequency sections because of the limited open-loop gain of
feedback, the state-variable, and its close cousin, the biquad. They the op amp at high frequencies. A guideline is that the open-loop
are the most common and are relevant here. More complete gain of the op amp should be at least 20 dB (i.e., ×10) above the
information on the various topologies is given in the References. amplitude response at the resonant (or cutoff) frequency, including
the peaking caused by the Q of the filter. The peaking due to Q
Sallen-Key, Low-Pass Filter will have an amplitude of magnitude A0:
The widely used Sallen-Key configuration, also known as a
voltage-controlled voltage source (VCVS), was first introduced (5)
in 1955 by R. P. Sallen and E. L. Key of MIT’s Lincoln Labs
(see Reference 3). Figure 10 is a schematic of a Sallen-Key, where H is the gain of the circuit.
second-order, low-pass filter. One reason for this configuration’s OUT
popularity is that its performance is essentially independent of R +
C
the op amp’s performance because the amplifier is used primarily
R R
as a buffer. Since the follower-connected op amp is not used for IN
+
voltage gain in the basic Sallen-Key circuit, its gain-bandwidth C
requirements are not of great importance. This implies that, for a
given op amp bandwidth, a higher-frequency filter can be designed
using this fixed (unity) gain, as compared to other topologies Figure 12. 2-pole, multiple-feedback (MFB), low-pass filter.
that involve the amplifier’s dynamics in a variable feedback loop.
The signal phase is maintained through the filter (noninverting The multiple-feedback filter inverts the phase of the signal. This is
configuration). A phase shift-vs.-frequency plot for a Sallen-Key, equivalent to adding 180° to the phase shift of the filter itself. The
low-pass filter with Q = 0.707 (or a damping ratio, α = 1/Q of variation of phase vs. frequency is shown in Figure 4 (right axis). This
1.414—Butterworth response) is shown in Figure 4 (left axis). To will be referred to as the inverted, second-order, low-pass response.
simplify comparisons, this will be the standard performance for Of interest, the difference between highest- and lowest-value
the second-order sections to be considered here. components to achieve a given response is higher in the multiple-
feedback case than in the Sallen-Key realization.
+

IN OUT
R R C
The Multiple-Feedback (MFB), High-Pass Filter
C
+ Comments made about the multiple-feedback, low-pass case apply
to the high-pass case as well. The schematic of a multiple-feedback,
high-pass filter is shown in Figure 13, and its ideal phase shift vs.
Figure 10. 2-pole, Sallen-Key, low-pass filter. frequency is shown in Figure 5 (right axis). This was referred to
as the inverted, second-order, high-pass response.
The Sallen-Key, High-Pass Filter
To transform the Sallen-Key low-pass into a high-pass OUT
con f ig u rat ion, t he capacitor s a nd t he resistor s i n t he C
+ R
frequency-determining network are interchanged, as shown
+

in Figure 11, again using a unity-gain buffer. The phase IN

shift vs. frequency is shown in Figure 5 (left axis). This is C R C

the in-phase, second-order, high-pass response.

+

IN OUT
C + R Figure 13. 2-pole, multiple-feedback (MFB), high-pass filter.
C

R This type of filter may be more difficult to implement stably at

high frequencies because it is based on a differentiator, which,
like all differentiator circuits, maintains greater closed-loop gain
Figure 11. 2-pole, Sallen-Key, high-pass filter. at higher frequencies and tends to amplify noise.

Analog Dialogue 41-10, October (2007) 

State-Variable The phase shift vs. frequency of the LOWPASS1 section will be
A state-variable realization is shown in Figure 14. This the in-phase, second-order, low-pass response (see Figure 4, left
configuration offers the most flexible and precise implementation, axis). The LOWPASS2 section will have the inverted second-
at the expense of many more circuit elements, including three order response (see Figure 4, right axis). The HIGHPASS section
op amps. All three major parameters (gain, Q, and ω0) can be has a phase shift that inverts (see Figure 5, right axis).
adjusted independently; and low-pass, high-pass, and band-pass
outputs are available simultaneously. The gain of the filter is also CONCLUSION
independently variable. We have seen that the topology used to build a filter will have
an effect on its actual phase response. This may be one of the
Since all parameters of the state variable filter can be adjusted factors used in determining the topology used. Table 1 compares
independently, component spread can be minimized. Also, the phase-shift ranges for the various low-pass filter topologies
mismatches due to temperature and component tolerances are discussed in this article.
minimized. The op amps used in the integrator sections will have
the same limitations on op amp gain-bandwidth as described in Table 1. Low-pass-filter topology phase-shift ranges.
the multiple-feedback section. LOW-PASS FILTERS

R
LOWPASS SINGLE PHASE
HIGHPASS FILTER TOPOLOGY PHASE VARIATION
R R
IN R R Single-Pole, Passive In-Phase 0° to –90°
+

C C Single-Pole, Active Inverted 180° to 90°

2-Pole, Sallen-Key In-Phase 0° to –180°
R R
BANDPASS 2-Pole, Multiple Feedback Inverted 180° to 0°
2-Pole, State Variable Inverted 180° to 0°
Figure 14. 2-pole, state-variable filter.
2-Pole, Biquad Lowpass1 In-Phase 0° to –180°
The phase shift vs. frequency of the low-pass section will be an 2-Pole, Biquad Lowpass2 Inverted 180° to 0°
inverted second-order response (see Figure 4, right axis) and
the high-pass section will have the inverted high-pass response Similarly, the various high-pass topologies are compared in
(see Figure 5, right axis). Table 2.
Table 2. High-pass-filter topology phase-shift ranges.
Biquadratic (Biquad)
A close cousin of the state-variable filter is the biquad (see Figure 15). HIGH-PASS FILTERS
The name of this circuit, first used by J.Tow in 1968 (see Reference 6), SINGLE PHASE
and later by L. C. Thomas in 1971 (see Reference 5), is based on the FILTER TOPOLOGY PHASE VARIATION
fact that the transfer function is a ratio of two quadratic terms. This Single-Pole, Passive In-Phase 90° to 0°
circuit is a slightly different form of a state-variable circuit. In this
configuration, a separate high-pass output is not available. However, Single-Pole, Active Inverted –90° to –180°
there are two low-pass outputs, one in-phase (LOWPASS1) and one 2-Pole, Sallen-Key In-Phase 180° to 0°
out-of-phase (LOWPASS2). 2-Pole, Multiple Feedback Inverted 0° to –180°
R1 2-Pole, State Variable Inverted 0° to –180°
+

R2 R3 R4
2-Pole, Biquad Inverted 0° to –180°
C1 C2

IN
R6 R5 The Variation of Phase Shift with Q
The second-order responses above have all used a Q of 0.707.
BANDPASS LOWPASS1 LOWPASS2
Figure 17 shows the effect on phase response of a low-pass filter
(the results for high-pass are similar) as Q is varied. The phase
Figure 15. Standard biquad, 2-pole section. responses for values of Q = 0.1, 0.5, 0.707, 1, 2, 5, 10, and 20 are
plotted. It’s worth noting that the phase can start to change well
With the addition of a fourth amplifier section, high-pass, notch
below the cutoff frequency at low values of Q.
(low-pass, standard, and high-pass), and all-pass filters can be
0
realized. A schematic for a biquad with a high-pass section is Q = 20
shown in Figure 16.
LOWPASS
R –45
BANDPASS
R
+

R C
PHASE

IN Q = 0.1
R R R –90
+C

LOWPASS1 –135
R R

HIGHPASS
R R

–180
0.01 0.1 1 10 100
FREQUENCY

Figure 16. 2-pole biquad filter (with a high-pass section). Figure 17. Variation of phase shift as Q is varied.

 Analog Dialogue 41-10, October (2007)

Although not the subject of this article, the variation of amplitude As noted earlier, the SK section is noninverting, while the MFB
response with Q may also be of interest. Figure 18 shows the section inverts. Figure 20 compares the phase shifts of these three
amplitude response of a second-order section as Q is varied over fourth-order sections. The SK and the MFB filters have the same
above range. response because two inverting sections yield an in-phase response
The peaking that occurs in high-Q sections may be of interest (–1 × –1 = +1). The filter built with mixed topologies (SK and
when high-Q sections are used in multistage filters. While in MFB) yields a response shifted by 180° (+1 × –1 = –1).
theory it doesn’t make any difference in which order the sections 180
are cascaded, in practice it is typically better to place low-Q 135
sections ahead of high-Q sections so that the peaking will not 90
cause the dynamic range of the filter to be exceeded. Although 45
MFB – SK
this plot is for low-pass sections, high-pass responses will show 0
similar peaking. –45

PHASE
20 –90
Q = 20 –135 SK – SK
10
–180
0
–225
MFB – MFB
–10
–270
–20
AMPLITUDE

–315
Q = 0.1
–30 –360
0.01 0.1 1 10 100
–40 FREQUENCY

–50
Figure 20. Fourth-order phase response with
various topologies.
–60

–70
Note that the total phase shift is twice that of a second-order
section (360° vs. 180°), as expected. High-pass filters would have
–80
0.01 0.1 1 10 100
similar phase responses, shifted by 180°.
FREQUENCY This cascading idea can be carried out for higher-order filters, but
anything over eighth-order is difficult to assemble in practice.
Figure 18. Amplitude peaking in 2-pole filter as Q is varied.
Future articles will examine phase relationships in band-, notch-
Higher-Order Filters (band-reject), and all-pass filters.
Transfer functions can be cascaded to form higher-order responses.
When filter responses are cascaded, dB gains (and attenuations) References
add, and phase angles add, at any frequency. As noted earlier, 1. Daryanani, G. Principles of Active Network Synthesis and Design.
multipole filters are typically built with cascaded second-order J. Wiley & Sons. 1976. ISBN: 0-471-19545-6.
sections, plus an additional first-order section for odd-order filters. 2. Graeme, J., G. Tobey, and L. Huelsman. Operational Amplifiers
Two cascaded first-order sections do not provide the wide range Design and Applications. McGraw-Hill. 1971. ISBN 07-064917-0.
of Q available with a single second-order section. 3. Sallen, R. P., and E. L. Key. “A Practical Method of Designing
A fourth-order filter cascade of transfer functions is shown in RC Active Filters.” IRE Trans. Circuit Theory. 1955. Vol. CT-2,
Figure 19. Here we see that the filter is built of two second- pp. 74-85.
order sections. 4. Thomas, L. C. “The Biquad: Part II—A Multipurpose Active
Filtering System.” IEEE Trans. Circuits and Systems. 1971. Vol.
CAS-18. pp. 358-361.
FILTER 1 AMPLIFIER 1 FILTER 2 AMPLIFIER 2 5. Thomas, L. C. “The Biquad: Part I—Some Practical Design
Considerations.” IEEE Trans. Circuits and Systems. 1971. Vol.
CAS-18. pp. 350-357.
Figure 19. Cascaded transfer functions for a 4-pole filter. 6. Tow, J. “Active RC Filters—A State-Space Realization.” Proc.
IEEE. 1968. Vol. 56. pp. 1137-1139.
Figure 20 shows the effect on phase response of building a
fourth-order filter in three different ways. The first is built with 7. Van Valkenburg, M. E. Analog Filter Design. Holt, Rinehart &
two Sallen-Key (SK) Butterworth sections. The second consists Winston. 1982.
of two multiple-feedback (MFB) Butterworth sections. The 8. Williams, A. B. Electronic Filter Design Handbook. McGraw-
third is built with one SK section and one MFB section. But just Hill. 1981.
as two cascaded first-order sections don’t make a second-order 9. Zumbahlen, H. “Analog Filters.” Chapter 5, in Jung, W.,
section, two cascaded second-order Butterworth sections do not Op Amp Applications Handbook. Newnes-Elsevier (2006).
equal a fourth-order Butterworth section. The first section of a (Original chapter from ADI Seminar Notes is available online.)
Butterworth filter has an f0 of 1 and a Q of 0.5412 (α = 1.8477). The 10. Zumbahlen, H. Basic Linear Design. Ch. 8. Analog Devices
second section has an f0 of 1 and a Q of 1.3065 (α = 0.7654). Inc. 2006. (Available soon).

Analog Dialogue 41-10, October (2007)