LAB 2: Measuring Capacitance

1. Introduction
Capacitance
A voltage difference V applied between two conductors will induce opposite charges +Q and –Q
on each. Capacitance is the ratio of the induced charge to the applied voltage, C = Q / V , and is
determined only by the shape of the conductors and the spacing between them. Capacitors are
used to temporarily store electric charge and energy. They are also used to condition signals in
analog and digital circuits.

In today’s lab we will construct a system for measuring capacitance and use this to investigate
two situations. First, you will measure the capacitance of a pair of circular plates separated by
thin sheets of polycarbonate and use your results to find the dielectric constant of this material.
Second, you will determine the capacitance of a pair of rectangular plates as a function of the
separation between them. At small plate separations we expect the capacitance to follow the
simple formula C = Aε 0 / d , but for separations comparable to the width of the plates, the
situation becomes more complex. In class you will use numerical techniques to calculate the
capacitance in this case. We reference an analytical solution below.

Determining Capacitance from AC Impedance

We can measure the capacitance of an element by applying an AC
voltage across it while we measure the AC current through it. A
short detour into AC circuit theory explains how this works: vin
Z
In AC circuits the current and voltage vary sinusoidally with time
at some frequency f = ω/2π. For example, in the circuit below a
voltage source produces an AC voltage, and an AC current flows
through a component with impedance Z.

Vin = V0 cos(ωt)
I = I0 cos(ω t + φ )

__________________________________________________________ 1
P443 L2
It is convenient to work with the complex quantities
jωt
V˜ = V0 e , I˜ = I0 e
j (ωt +φ )
. ( j = −1 ).

The complex current and voltage are related by a form of Ohm's law, V˜ = ZI˜ .

The impedance Z of resistors, capacitors and inductors are given by the following formulas, and
we note that the impedance of capacitors and inductors depend on frequency.

Resistor Z R= R resistors are frequency independent

Capacitor ZC = (jωC)-1 capacitors block DC
Inductor ZL= jωL inductors block high frequency

We can combine the impedance of different components in the same way as we add resistances
in series and parallel. In the end, we find the actual current and voltage by taking the real part of
the complex quantities.

The circuit we are studying can be treated as a

perfect AC voltage source connected across an
ideal resistor and capacitor in series. The
capacitance is due to the parallel plates (plus
stray capacitance from other parts of the circuit).
The resistance includes the resistance of the
wires and the output impedance of the AC
voltage source.

If there were no resistive component to the circuit, then the ratio of the current to voltage would
be simple: I!C / V!C = 1 / Z C = jω C , and the actual measured current would be:

I C (t) = ω CV0 sin(ω t) . (R=0) (1)

A measurement of the amplitude of the current would determine C.

If the resistance of the circuit is not negligible, then the total impedance of the circuit is the sum
of the impedance of the capacitor and resistor:

Z = Z R + ZC = R + 1 jω C .

The ratio of the current to the voltage is then

⎛ 1 ⎞ 1
I! / V! = 1 / Z = 1 / ⎜ R + ⎟ = jω C .
⎝ jω C ⎠ 1+ jRCω

__________________________________________________________ 2
P443 L2
If the resistance is a small perturbation to the circuit (ωRC <<1) we can approximate this term by
expanding the fraction and keeping only terms to first order:

I! / V! ≈ jω C (1− jRCω ) = RC 2ω 2 + jω C (2)

The actual current would then be:

I(t) = (ω 2 RC 2V0 ) cos(ω t) + (ω CV0 ) sin(ω t) . (3)

Our measurement will use a lock-in amplifier that can separately measure the cosine-component
and sine-component of the signal. A measurement of the sine-component of the current
determines C.

2. Apparatus
The apparatus consists of the physical capacitors and the electronics. The physical capacitors
consist of parallel circular or rectangular plates mounted on rails so that the plate separation can
be varied. The electronics include a function generator as an AC source and a lock-in amplifier
as an AC current meter. An oscilloscope allows the operator to monitor the signal in real time.

Oscillo
-scope

CH 1 CH 2.

Monitor

Lock-in Amp.
Input A/I Ext./Ref

CH 1
CH 2.

Function Generator
Physical Capacitor
GWINSTEK AFG-2225 Function Generator: The function generator can generate two
independent periodic waveforms at frequencies up to 25MHz. The output waveform can be
sinusoidal, square wave, pulse, or linear ramp. Alternately, the user can program an arbitrary
waveform shape. Today we will use sinusoidal outputs at frequency f = 10KHz for CH1 and
CH2. CH1 will be the voltage source for our experiment. CH2 will be used to trigger our
electronics.

__________________________________________________________ 3
P443 L2
Stanford Research SR-830 Lock-in Amplifier: A Lock-in Amplifier acts as an AC filter tuned
to the frequency of the signal with an arbitrarily narrow pass-band. Such a filter will reject any
noise that is not within the pass-band and allow the signal to be measured. A typical example
might use a center frequency f =10KHz and a bandwidth Δf = 1Hz. Lock-in amplifiers can often
recover small signals even when the noise is hundreds of times larger than the signal.

A Lock-in Amplifier requires a signal at a fixed frequency and a reference. The reference is a
“clean” AC voltage at the frequency of the signal. The Lock-in internally generates an AC
voltage at the frequency of the reference. The signal is then multiplied by the AC voltage and
the product is averaged over time. Noise at other frequencies averages to zero because sine-
functions of different frequencies are orthogonal.

Two channel Lock-in Amplifiers like the SR-830 also measure the component of the signal at the
reference frequency but 90º out of phase with the reference. Mathematically these two
operations can be represented:

T T

Vxout = 2
T ∫ Vs (t)cos(ω t)dt ,
0
Vyout = 2
T ∫ V (t)sin(ω t)dt .
0
s

The in-phase component ( Vxout ) is just the Fourier Cosine Transform of the signal, and the out-
of-phase component ( Vyout ) is the Fourier Sine Transform. Today we will use the Lock-in
amplifier to separately measure the cosine-component and sine-component of the AC current
through a capacitor.

3. Procedure
I. SET-UP
1. Connect the electronics using the figure above as a guide. Connect the PASCO physical
capacitor with circular plates. Set the distance between the plates to 1.0 cm.

2. Set up the function generator as follows:

Waveform Frequency Amplitude Operate

CH. 1 sinusoidal 10KHz 10mV RMS ON
CH. 2 sinusoidal 10KHz 1V RMS ON

3. Measure the CH1 output voltage with a DMM set to AC volts. Surprised?

__________________________________________________________ 4
P443 L2
4. Set up the Lock-in Amplifier as follows:

Input Sensitivity Time Constant Mode

I (1e-6) 100mV 300ms CH1: X CH2: Y

5. Record the current shown in CH2. According to the theory above

this current is determined by the capacitance I y = 2π fCV0 .
Determine the capacitance from the measured current.

6. Compare your result to the expected capacitance for an ideal

parallel plate capacitor, C0 = Aε 0 d . The circular plates have
radius R = 10cm.

II. DIELECTRIC CONSTANT OF POLYCARBONATE

1. Measure the thickness of each of the 12” x 12” polycarbonate
sheets with a caliper.

2. Place the thinnest polycarbonate sheet between circular plates of

the PASCO physical capacitor. Gently press the plates together
(touching only the plastic parts, not the metal plates) while you
record the current shown in CH2.

3. Repeat for each of the other polycarbonate sheets.

III. RECTANGULAR CAPACITOR

1. Replace the PASCO physical capacitor with the rectangular plate
physical capacitor.

2. Record the current shown in CH2 for a series of plate separations.

I suggest you take measurements for about 15 plate distances from
5mm to 150mm. Start with a small distance increment and
increase the increment as you get to larger plate separations.
Check that the plates are parallel. Make sure there are no
conductors anywhere near the plates when you take a
measurement. You are a conductor!

__________________________________________________________ 5
P443 L2
4. Analysis
Set-up an excel spreadsheet to convert the current measurements to capacitance values.
This will save you a lot of time in the rest of the analysis.

I. DIELECTRIC CONSTANT OF POLYCARBONATE

1. Find the capacitance values for the PASCO physical capacitor with each polycarbonate
sheet.

2. Plot the capacitance versus sheet thickness.

3. For parallel-plate capacitors filled with a dielectric material the predicted capacitance is
C = Aε r ε 0 / d . Fit your data to find the relative dielectric constant of polycarbonate ε r .

II. RECTANGULAR CAPACITOR

1. Find the capacitance values Cexp for the Rectangular physical capacitor for each plate
separation you investigated.

2. Calculate C0 = Aε 0 / d for the plate separations you measured. Plot Cexp and C0 versus d
together on one graph.

3. Find the Normalized Capacitance CPN = Cexp / C0 as a function of the Aspect Ratio
b = d / W from your data.

3. In the paper “Form and Capacitance of Parallel-Plate Capacitors” by Hitoshi Nishiyama

and Mitsunobu Nakamura (IEEE TRANSACTIONS ON COMPONENTS, PACKAGING, AND
MANUFACTURING TECHNOLOGY-PART A, VOL. 17, NO. 3. SEPTEMBER 1994), the authors
present calculations of the capacitance of physical capacitors of different shapes. Table 1
summarizes their results (next page). Our system is most similar to the Strip Capacitor
(Rectangular parallel plates with L >> W)

4. Plot your values for CPN versus b together with the values from Nishiyama and
Nakamura for the Strip Capacitor from the table above.

__________________________________________________________ 6
P443 L2
 __________________________________________________________ 7
P443 L2