RC Filter Networks Lab Report

Robert Clancy 04329741

December 8, 2005

Abstract
The aims of this experiment are to examine RC circuits as low-pass
and high-pass filters to source voltages of sinusoidal and square wave
signals. Their responses to different frequencies are investigated and
the ‘half power point’ is measured.

1 Theory
In an RC circuit the total impedance is

Z = R − iXC
Z = |Z| exp(iθ)

where XC = 1/ωC and θ = arctan(−XC /R).

The voltage drop across the resistor and capacitor are given respectively
by
R
VR = Vin
R − iXC
−iXC
VC = Vin
R − iXC

1.1 High-pass
In a high-pass RC filter circuit the output voltage is taken across the resistor.
From the equation above the magnitude of the output voltage is

|Vout | R
=q
|Vin | R2 + XC2
|Vout | R
=p
|Vin | R + (1/ωC)2
2

|Vout | ωCR
=√
|Vin | 1 + ω 2 C 2 R2

1
so |Vout | increases with increasing ω to a maximum of |V in |. It allows high
frequencies through and attenuates low frequencies. Clearly it is correctly
called a high-pass circuit.
The phase difference between the input voltage and the output voltage
is (from the equation above)
 
XC
θhigh = arctan
R
 
1
θhigh = arctan
ωCR

so the output voltage leads the input voltage by θ high and this phase differ-
ence decreases with increasing frequency.

1.2 Low-pass
In a low-pass circuit the output voltage is taken across the capacitor so the
magnitude of the output voltage is in this case

|Vout | XC
=q
|Vin | R2 + XC2
|Vout | 1/ωC
=p
|Vin | R + (1/ωC)2
2

|Vout | 1
=√
|Vin | 1 + ω 2 C 2 R2
clearly |Vout | decreases with increasing ω so it attenuates high frequencies
and passes low frequencies.
VR

θhigh

θlow
PSfrag replacements

VC Vin

Figure 1: Phasor diagram showing the voltage drop across the capacitor,
the resistor, and the source voltage. V C = IXC and VR = IR so clearly
tan(θhigh ) = XC /R and θlow = θhigh − π2 .

2
 The phase difference between the input voltage and the output voltage
is
 
−R
θlow = arctan
XC
θlow = arctan (−ωCR)

the output voltage lags the input voltage by θ low and increases with increas-
ing frequency to a maximum of π2 .
These results can easily be summarized in a phasor diagram, see figure 1.

1.3 Integrating and differentiating circuits

Using Kirchoff’s laws we can show that the voltage across a capacitor in an
RC circuit with a constant voltage source V 0 is

VC = V0 [1 − exp(−t/τ )]

where τ = RC is the time constant of the circuit. If t = τ then

VC
= 1 − exp(−1)
V0
VC
≈ 0.63
V0
For a general source voltage Vin the sum of VR = RI and VC = Q/C is
Vin . Since I = dQ/dt we have

VR = RI
d(VC C)
VR = R
dt
dVC
VR = RC
dt

and similarly for VC

1
Z
VC = VR dt
RC
If we apply a square wave to this circuit with period T then we have two
distinct cases. If T  RC then the capacitor charges fully every period and
VC ≈ Vin , also
dVC
VR ∝
dt
dVin
VR ∝
dt

3
If T  RC then the capacitor does not have time to charge and so V R ≈ Vin
and so
Z
VC ∝ Vin dt

In this way we have simple integrating and differentiating circuits.

2 High-Pass Filter
2.1 Method and Results
The high-pass RC filter circuit was set up as shown in figure 2. The peak-
to-peak values of the output voltage, V out , and the time difference, t, in
seconds was measured for different frequencies and a constant V in . The
phase difference was calculated using θ high = 2πf t. Since the data would
be plotted on a logarithmic scale the frequency was varied according to a
logarithmic scale.

f /Hz Vout /V t/ms θhigh /rad

10±.02 .11±.001 25±2 1.6±.1
14±.04 .15±.001 17±1 1.50±.09
19±.03 .20±.002 12±1 1.4±.1
26±.04 .274±.002 9.5±.5 1.55±.08
36±.04 .376±.002 6.5±.5 1.5±.1
49±.05 .512±.004 5±.2 1.54±.06
68±.1 .704±.004 3.4±.2 1.45±.09
94±.2 .976±.004 2.5±.1 1.48±.06
130±.1 1.34±.01 1.8±.1 1.47±.06
178±.4 1.86±.02 1.25±.05 1.40±.06
245±.3 2.52±.01 .85±.05 1.31±.08
337±.4 3.38±.02 .6±.05 1.3±.1
464±.5 4.48±.04 .4±.02 1.17±.06
640±.6 5.72±.04 .25±.01 1.01±.01
880±2 7.04±.04 .16±.01 .88±.06
1210±5 8.32±.01 .09±.005 .68±.04
1670±10 9.60±.01 .05±.005 .52±.06
2300±20 10.1±.01 .028±.002 .40±.03
3200±10 10.4±.1 .015±.001 .30±.02
4500±10 10.7±.1 .008±.0005 .23±.01
6000±20 10.8±.1 .0045±.0005 .17±.02

Vin = 11.1 ± .05V

4
 C = 0.022µF

R = 6.8kΩ Vout
PSfrag replacements V
in

Figure 2: High-pass RC circuit. The output is taken over the resistor.

2.2 Discussion
Figure 3 shows the dependence of Vout on the frequency of the input sinu-
soidal voltage, f . The error bars have been left out because they were too
small to see. It is convenient to use a logarithmic scale on the x-axis be-
cause it allows us to look at a wide range of f while keeping the numerical
values low. This is an increasing function, tending towards zero. This trend
matches with the theory for the high-pass circuit since V out tends to Vin for
higher frequencies.

0
-0.2
-0.4
-0.6
log10 (Vout /Vin )

-0.8
-1
-1.2
-1.4
-1.6
-1.8
-2
0.5 1 1.5 2 2.5 3 3.5 4
log10 f

Figure 3: The voltage increases with increasing frequency for a high-pass

circuit tending towards Vin which would be zero on this graph.

5
 The half power point is the point at which the output power is a half
the√input power or since P ∝ V 2 the point at which the output voltage is
1/ 2 times the input voltage. This is called the 3 dB point because using
the decibel scale

I = 10dB log 10 (Pin ) − 10dB log 10 (Pout )

= 10dB (log 10 (Pin ) − log10 (Pin /2))
= 10dB log 10 (2)
≈ 3dB

We can work out the frequency at the half power point from the theory.
1 ωCR
√ =√
2 1 + ω 2 C 2 R2
ωCR = 1
1
f=
2πCR
1
f=
2π × 0.022 × 10−6 × 6.8 × 103
f = 1064Hz

The half power point on the graph was found to be log 10 (f ) = 3.04±0.01
so f = 1100 ± 30Hz. so the measured value for the half power point agrees
well with the theoretical value.

1.8
1.6
1.4
1.2
θhigh /rads

1
0.8
0.6
0.4
0.2
0
0.5 1 1.5 2 2.5 3 3.5 4
log10 f

Figure 4: The phase difference in a high-pass circuit is a decreasing function

tending towards zero.

6
 Figure 4 shows the variation of the phase angle θ high with f . Again
the error bars on the x-axis were left out because they were small. The
phase angle at the half power point was determined from the graph to be
θhigh = .74 ± .05. At the half power point we have

ωCR = 1

so

θhigh = arctan(1)
θhigh = .785

the measured phase angle at half power point is within experimental of the
theoretical value. The input was leading the output throughout the high-
pass experiment.

3 Low-pass Filter

R = 680Ω

C = 0.22µF Vout
PSfrag replacements V
in

Figure 5: Low-pass RC circuit. The output is taken over the capacitor.

The low-pass RC filter circuit was set up as shown in figure 5. The

frequency and phase shift at the half-power point were measured.
The frequency at the half-power point was 1.14 ± .03kHz. This is not
within experimental error of the previously calculated theoretical value of
1064Hz. It is an error of approximately 6%. The time difference was 110 ±
5µs which corresponds to a phase difference of λ low = .79 ± .06 rad. This is
within experimental error of the theoretical value of .785. In this case the
output was leading the input voltage.

7
4 Square wave response of filter circuits
A square wave was applied to the high-pass and low-pass circuits shown in
figures 2,5 respectively.
For the period T  τ , the time constant, the output voltage shows
exponential growth. The time constant was measured by checking the time
at which the output voltage was 0.63 times the input voltage. This was
found to be 151 ± 2µs. This is within experimental error of the calculated
value of
τth = CR = 0.22 × 10−6 × 680 = 149µs
The figures 6-8 show the three cases of T  τ , T ≈ 2τ , and T  τ for
the high-pass and low-pass circuits.
Figure 6(a) shows the high-pass circuit for T = 7.15ms. In this case
T  RC so the circuit acts as a differentiating circuit. If we compare
figure 6(a) with figure 6(b) we see that the high-pass circuit shows the slope
of the output from the low-pass circuit. Also we note that the output voltage
from the low-pass circuit shown in figure 6(b) approximates the input square
voltage.
Figures 7(a) and 7(b) show the case for T ≈ 2RC. We can see the
exponential growth and decay in figure 7(b) and the corresponding output
over the resistor in figure 7(a). Again we note that the high-pass circuit
displays the slope of the low-pass circuit and conversely the low-pass circuit
shows the area under the high-pass circuit.
Figures 8(a) and 8(b) shows when T = 30.40µs so T  RC. The output
across the high-pass filter approximates the square wave input voltage and
so the output of the low-pass filter is approximately the integral of the square
wave input. This is the integrating circuit.
These results can be qualitatively predicted if we know that the Fourier
series of the square wave input is
1 1
x(t) = sin(ωt) + sin(3ωt) + sin(5ωt) + . . .
3 5
For T  RC i.e. a low frequency input the high-pass filter will only
allow through the high-frequency parts of the Fourier series, which gives us
the wave shown in figure 6(a). The low-pass filter circuit on the other hand
will allow through most of the series and so the output will approximate the
input, as seen in figure 6(b).
If T  RC the high-pass filter allows through all the components of the
Fourier series and so the high-pass output approximates the input square
wave. The low-pass filter circuit in this case attenuates the higher frequen-
cies greatly and the lower frequencies that pass produce the wave shown in
figure 8(b).

8
5 Conclusions
The half power point of the high-pass filter was found to be f = 1100±30Hz
and the the output was lagging the input at this frequency by θ high =
.74±.05. These are both within experimental error of the theoretical values.
The half power point of the low-pass filter circuit was measured to be
f = 1140 ± 30Hz which is an error of 6% compared to the theoretical value
of 1064Hz. The output was leading the input by θ low = .79 ± .06rads.
The time constant of the low-pass circuit was measured to be τ = 151 ±
2µs. This is within experimental error of 149µs. The RC filter circuits were
also investigated as simple integrating and differentiating circuits.

References
[1] W.J. Duffin. Electricity and Magnetism. McGraw-Hill.