The Time Constant of An RC Circuit: 1 Objectives
The Time Constant of An RC Circuit: 1 Objectives
The Time Constant of An RC Circuit: 1 Objectives
1 Objectives
1. To determine the time constant of an RC Circuit, and
2 Introduction
What the heck is a capacitor? Its one of the three passive circuit elements: the resistor
(which weve already met), the capacitor (which stores energy in electric fields), and the
inductor (which stores energy in magnetic fields, and is the main subject a few weeks from
now). Also known as condensers, capacitors store energy an electric field by separating
positive and negative charges on opposing terminals known as plates - which may or may
not actually be plate-like. Capacitors trace their lineage to the middle of the Eighteenth
century and the invention of the Leyden jar. For forty years, the biggest names in natural
philosophy spent time in the study of charge storage and improving capacitors, including
Volta (for whom we name the volt) and Franklin (of American Founding Father fame).
In this lab, we will study the behavior of capacitors while we add or remove charge from
the plates. Although we wont touch on it in this lab, RC circuits form the foundation of the
modern electronics that underlie a whole host of technologies, including radio, computers,
environmental and medical sensors, and - not least of all - the touch screens in cell phones
and the touch pads of laptops.
3 Theory
We normally write Ohms Law for the resistor as V (t) = I(t)R. But what is that current
I(t)? In the circuits here (see Figure 1), its just the rate at which charge passes through
the resistor. Since all the charge that runs through the resistor ends up going from or to the
capacitor, q(t), it must also be the case that
dq(t)
I(t) = ,
dt
where q(t) is the time dependent charge on the capacitor, with the sign depending on whether
the charge on the capacitor is increasing or decreasing. For this circuit, we could instead
1
Figure 1: The switching circuit used to discuss charging and discharging a capacitor.
q(t) = CV (t) .
The constant C is called the capacitance, and is measured in Farads (named for Michael
Faraday, whose research output we will learn about later this semester) where F = C/V.
Since capacitors are used to store charge, we must find a way to change the charge state
on the capacitor. Lets discuss the circuit in Figure 1. This circuit has three states: 1. The
steady state, where the switch is open, no current flows in the resistor, and the charge state
on the capacitor is constant, 2. the charging state, where the battery or power supply is
connected to the capacitor and adds charge to the capacitor, and 3. the discharging state,
where the battery is disconnected, the two plates of the capacitor are connected to each
other through the resistor, which removes charge from the capacitor.
We can analyze the dynamic states of the circuit using Kirchoffs Rules1 :
1. The Loop rule (energy conservation) requires the sum of all voltages drops around a
closed loop to vanish, and
2. The Junction rule (charge conservation) requires the sum of all currents into a junction
to vanish.
Lets apply these rules, first to the discharging and then to the charging state:
1. Discharging: In the discharging case, the current will flow off the capacitor in a counter
clockwise direction (why?). When we close that switch, Kirchoffs rules become
Vc VR = 0
dq(t) q(t)
=0.
dt RC
1
As long as the frequencies arent so high that we cant use the lumped element approximation; talk to
your instructor if you want to know more.
2
Again, the rate is negative, because the capacitor is discharging, not charging. The
method of solution for this equation is given in Appendix A:
assuming that the initial voltage across the capacitor is Vs . This discharge curve is
plotted in Figure 2.
2. Charging: In the charging case, the current flows clockwise from the battery (at voltage
Vs ), through the resistor (at VR ), and across the capacitor Vc ; we want to solve for Vc
as a function of time. Kirchoffs Rules tell us
Vs VR Vc = 0
q
Vs IR
C
But here, the current through the resistor depends on the rate at which charge is chang-
ing on the capacitor; they have the same sign here since the charge on the capacitor is
increasing
dq(t) q(t)
Vs R =0
dt C
dq(t) q(t) Vs
+ = .
dt RC R
We can solve this differential equation too for the time dependent voltage profile across
the capacitor; see Appendix A. If we start with a completely discharged capacitor, the
voltages across the resistors and capacitors vary as
Vc (t) = Vs 1 et/RC
VR (t) = Vs et/RC .
As the capacitor charges, the voltage across - and hence the charge on - the capacitor
rises exponentially, while the voltage across - and hence the current through - the
resistor will fall exponentially; see Figure 2.
The quantity RC - which appears in the argument of the exponential - is known as the
time constant of the system; it has units of time (hence the name), and determines the time
interval over which voltages, charges, and currents change in the circuit. The time constant
can be tuned by modifying either R or C. In practice, more resistor than capacitor values
are commercially available, so its usually easier to tune the resistor values.
Given a known resistance we can measure the time constant of the RC circuit, and
algebraically determine the capacitance C. When discharging the capacitor,
Vc (t) = Vs et/RC .
3
Capacitor charging and discharging curves
1
Discharging
Charging
0.8
0.6
Vc(t)/Vs
0.4
0.2
0
0 1 2 3 4 5
t/RC
Figure 2: The capacitor charging and discharging curves. The vertical blue line is the half
life point of the charging and discharging timeline.
We can easily measure and use the half-life T1/2 of the discharge: T1/2 is the time it takes for
the voltage to fall by half. Substituting into the above equations and solving for T1/2 :
Vs
= Vs eT1/2 /RC
2
1 T1/2
ln =
2 RC
T1/2 = (C ln 2) R .
If we allow R to vary, then this equation defines a line with abscissa R, ordinate T1/2 , and
slope C ln 2. If we measure multiple (R, T1/2 ) pairs, we can plot those data points, fit them
to a line, and extract the slope - and hence the capacitance.
4 Procedures
You should receive an oscilloscope, a function generator, a multimeter, one unknown capac-
itor, a decade capacitance box, and a decade resistance box.
4
Prototypical Oscilloscope Trace
Trace A
Trace B
1
0.8
0.6
Vc(t)/Vs
0.4
0.2
0 1 2 3 4 5
t/RC
Figure 3: The left-hand figure is the circuit used to measure the time constant of an RC
circuit, while the right-hand figure shows the Oscilloscope traces.
If the period of the square wave Ts is much less than the time constant = RC (Ts ),
then the capacitor will start discharging before it has sufficient time to acquire the maximum
charge q0 , making measurement of the time constant difficult. What will this look like on
the oscilloscope? If Ts is much more than (Ts ), you will not be able to fit both the
charging and discharging portions of the capacitor voltage on the oscilloscope trace at the
same time. Why?
1. Construct the circuit shown in Figure 3a.
2. Perform the initial turn-on procedure for the oscilloscope (from the Oscilloscope Lab).
To start, set both channels to 0.2 V/div. Make sure that the ground of both traces is
at the same position on the screen.
3. Start by setting the frequency of the square wave to about 100 Hz and the output
voltage to 0.5 V.
4. Now we have to set R and C. Adjust R to about 8 k, and C to about 0.1 F. What
is the time constant of this combination?
5. Adjust the frequency of the square wave, and the time and voltage settings of the
oscilloscope until you obtain traces similar to Figure 3b. Measure and record R, C, f ,
V , T1/2 , and the oscilloscope settings.
6. Measure the voltage of the decay curve above ground at a number of points using the
oscilloscope cursors; similarly, measure the height of the charging curve at a number
of points. You will use these to map out the charging/discharging curves, and fit for
the time constant.
7. Repeat for a second set of R and C.
5
1. Replace the known capacitor with the unknown capacitor in the circuit.
2. Set the resistance to about 4 k, and make the necessary adjustments to the oscillo-
scope settings to again obtain the appropriate display on the oscilloscope.
4. Repeat for four or five values of R; you need enough points to fit a line through your
(R, T1/2 ) pairs.
5. Use your multimeter to directly measure the capacitance, for comparison to your ex-
tracted value.2
A Derivation of Solutions
The differential equations in Section 3 are the simplest of all differential equations to solve:
first order, linear equations with constant coefficients. Here, well solve a slightly more
complicated equation (a first order linear equation with non-constant coefficients) in complete
generality, and use this solution to find the solutions were interested in. Consider the
following differential equation:
dy(x)
+ P (x)y(x) = Q(x) .
dx
This equation looks very much like the equations we found in Section 3 using Kirchoffs Rules.
We can solve this in the following way. First, well multiply both sides of this equation by a
new function, M (x), and well first solve for M (x).
dy(x)
M (x) + M (x)P (x)y(x) = M (x)Q(x) .
dx
Lets now assume that the left hand side is the expansion of the product rule of M (x)y(x):
d
(M (x)y(x)) = M (x)Q(x) .
dx
What must be true for this to hold? Well,
dM (x)
= M (x)P (x) .
dx
2
The multimeter uses an automated version of exactly this procedure to calculate C, based on a time
constant measurement across a precision internal resistor.
6
This is a trivial problem to solve: just separate and integrate:
R R R
M (x) = e P (x)dx+c
= ec e P (x)dx
= c0 e P (x)dx
,
where c is the constant of integration for this indefinite integral. Now, go a few lines back
up the page:
d
(M (x)y(x)) = M (x)Q(x) .
dx
We can solve this one trivially, too, just by integrating; lets make this a definite integral
between x0 and x. We get the left hand side from the Fundamental Theorem of Calculus
Z x
M (x)y(x) M (x0 )y(x0 ) = M (x)Q(x)dx ,
x0
where, since this is an indefinite integral, requires an integration constant c. Finally, then,
we can solve the general problem:
Z x
1
y(x) = M (x) M (x0 )y(x0 ) + M (x)Q(x)dx .
x0
Note that the integration constant in the definition of M (x) cancels here, so it doesnt really
matter what it was; lets just choose c0 = 1.
In the case of the charging RC circuit with zero initial charge, x = t, y(x) = q(t), x0 = 0,
y(x0 ) = 0, P (t) = 1/RC, and Q(t) = Vs /R. Integrating gives us M (t) = et/RC , giving us the
solution:
Z t
t/RC Vs
q(t) = e et/RC dt
0 R
Vs
= RCet/RC et/RC 1
R
= Vs C 1 et/RC ,
where weve used the integration constant to get the t = 0 value right (here, no charge on
the capacitor). Dividing both sides by C gives us the voltage across the capacitor at time t
Vc (t) = Vs 1 et/RC .
q(t) = Vs Cet/RC ,
where we assume the voltage across the capacitor at t = 0 is Vs . Again, divide by C to get
the voltage profile
Vc (t) = Vs et/RC .
7
Pre-Lab Exercises
Answer these questions as instructed on Blackboard; make sure to submit them before your
lab session!
8
Post-Lab Exercises
1. Determine the time constant of the RC circuit with known capacitance. Do this by
plotting your data of time versus the amplitude of the signal. You should fit this data
to an exponential, and extract . There are a number of ways to do this: (a) use a
mathematics package capable of sophisticated statistical analysis, such as R, Octave,
Matlab, Mathematica, Root, etc., or (b) using a spreadsheet, fit the data using an
exponential trendline analysis. Alternatively, you can transform your spreadsheet data
into a linear data set, and do a linear trendline analysis. Estimate the uncertainty
of this measured time constant. You should also estimate the time constant from the
measured T1/2 . Do these multiple experimental extractions agree within your estimated
uncertainty? Do they agree with the theoretical expectation given your measured R
and C?
2. Determine the unknown capacitance. Plot your (R, T1/2 ) data pairs, and fit the data
to a line. Extract the capacitance from the fitted line parameters. Estimate the
uncertainty of this extracted capacitance. You can also estimate the capacitance from
the individual (R, T1/2 ) pairs; estimate your uncertainties. Do these estimates agree
with the line fit estimate?
3. Discuss briefly whether you have met the objectives of the lab exercises.