www.lightandmatter.

com
 Copyright (c) 1999-2008 by B. Crowell and V. Roundy. This lab manual is subject to the Open Publication
License on page 182. If you do not agree to the license, then you do not have permission to copy the
manual. The lab manual is available for downloading from www.lightandmatter.com, and a copy of the
Open Publication License is also available at opencontent.org.

2
 Contents
1 Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 Free Fall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4 The Earth’s Gravitational Field . . . . . . . . . . . . . . . . . . . . . . . . 16
5 Newton’s Second Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6 Air Friction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7 Vector Addition of Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
9 Acceleration In Two Dimensions . . . . . . . . . . . . . . . . . . . . . . . . 24
10 Conservation Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
11 Conservation of Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
12 Conservation of Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . 34
13 Conservation of Momentum in Two Dimensions . . . . . . . . . . . . . . . . . . . 36
14 Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
15 The Moment of Inertia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
16 Absolute Zero . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
17 The Clement-Desormes Experiment . . . . . . . . . . . . . . . . . . . . . . . 46
18 The Pendulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
19 Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
20 Resonance (short version for physics 222) . . . . . . . . . . . . . . . . . . . . . 56
21 Standing Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
22 Resonances of Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
23 Static Electricity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
24 The Oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
25 The Speed of Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
26 Electrical Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
27 The Loop and Junction Rules . . . . . . . . . . . . . . . . . . . . . . . . . 82
28 Electric Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
29 Magnetism (Physics 206/211) . . . . . . . . . . . . . . . . . . . . . . . . . 88
30 The Dipole Field (Physics 222) . . . . . . . . . . . . . . . . . . . . . . . . . 92
31 The Earth’s Magnetic Field (Physics 222) . . . . . . . . . . . . . . . . . . . . . 96
32 Relativity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
33 The Charge to Mass Ratio of the Electron. . . . . . . . . . . . . . . . . . . . . 102
34 Energy in Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
35 RC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
36 LRC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
37 Faraday’s Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
38 Electromagnetism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
39 Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
40 Refraction and Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
41 Geometric Optics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
42 Two-Source Interference . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
43 Wave Optics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
44 Polarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
45 The Photoelectric Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
46 Electron Diffraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
47 The Hydrogen Atom . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
48 The Michelson Interferometer . . . . . . . . . . . . . . . . . . . . . . . . . 160
Appendix 1: Format of Lab Writeups . . . . . . . . . . . . . . . . . . . . . . 162

Contents 3
 Appendix 2: Basic Error Analysis . . . . . . . . . . . . . . . . . . . . . . . . 164
Appendix 3: Propagation of Errors . . . . . . . . . . . . . . . . . . . . . . . 170
Appendix 4: Graphing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
Appendix 5: Finding Power Laws from Data . . . . . . . . . . . . . . . . . . . . 174
Appendix 6: Using the Photogate . . . . . . . . . . . . . . . . . . . . . . . . 176
Appendix 7: Using a Multimeter . . . . . . . . . . . . . . . . . . . . . . . . 178
Appendix 8: High Voltage Safety Checklist . . . . . . . . . . . . . . . . . . . . 180
Appendix 9: Laser Safety Checklist . . . . . . . . . . . . . . . . . . . . . . . 182
Appendix 10: The Open Publication License . . . . . . . . . . . . . . . . . . . . 184

4 Contents
Contents 5
1 Interactions
Apparatus
single neodymium magnet . . . . . . . . . . . . . . . . 1/group
triple neodymium magnet . . . . . . . . . . . . . . . . 1/group
compass
triple-arm balance . . . . . . . . . . . . . . . . . . . . . . . . 2/group
clamp and 50-cm vertical rod for holding balance up
string
tape
scissors
heavy-duty spring scales
rubber stoppers
Goal
Form hypotheses about interactions and test them.
Introduction
Why does a rock fall if you drop it? The ancient
Greek philosopher Aristotle theorized that it was be-
cause the rock was trying to get to its natural place,
in contact with the earth. Why does a ball roll if you
push it? Aristotle would say that only living things
have the ability to move of their own volition, so the
ball can only move if you give motion to it. Aristo-
tle’s explanations were accepted by Arabs and Euro-
peans for two thousand years, but beginning in the
Renaissance, his ideas began to be modified drasti-
cally. Today, Aristotelian physics is discussed mainly
by physics teachers, who often find that their stu-
dents intuitively believe the Aristotelian world-view
and strongly resist the completely different version
of physics that is now considered correct. It is not
uncommon for a student to begin a physics exam
and then pause to ask the instructor, “Do you want
us to answer these questions the way you told us was
true, or the way we really think it works?” The idea
of this lab is to make observations of objects, mostly
magnets, pushing and pulling on each other, and to
figure out some of the corrections that need to be
made to Aristotelian physics.
Some people might say that it’s just a matter of
definitions or semantics whether Aristotle is correct
or not. Is Aristotle’s theory even testable? One
testable feature of the theory is its asymmetry. The
Aristotelian description of the rock falling and the
6 Lab 1 Interactions
ball being pushed outlines two relationships involv-
ing four objects:
According to Aristotle, there are asymmetries in-
volved in both situations.
(1) The earth’s role is not interchangeable with that
of the rock. The earth functions only as a place
where the rock tends to go, while the rock is an
object that moves from one place to another.
(2) The hand’s role is not analogous to the ball’s.
The hand is capable of motion all by itself, but the
ball can’t move without receiving the ability to move
from the hand.
If we do an experiment that shows these types of
asymmetries, then Aristotle’s theory is supported.
If we find a more symmetric situation, then there’s
something wrong with Aristotle’s theory.
Observations
The following important rules serve to keep facts
separate from opinions and reduce the chances of
getting a garbled copy of the data:
(1) Take your raw data in pen, directly into your lab
notebook. This is what real scientists do. The point
is to make sure that what you’re writing down is
a first-hand record, without mistakes introduced by
recopying it. (If you don’t have your two lab note-
books yet, staple today’s raw data into your note-
book when you get it.)
(2) Everybody should record their own copy of the
raw data. Do not depend on a “group secretary.”
(3) If you do calculations during lab, keep them on
a separate page or draw a line down the page and
keep calculations on one side of the line and raw
data on the other. This is to distinguish facts from
inferences.
Because this is the first meeting of the lab class,
there is no prelab writeup due at the beginning of
the class. Instead, you will discuss your results with
your instructor at various points.
A Comparing magnets’ strengths
To make an interesting hypothesis about what will
happen in part C, the main event of the lab, you’ll
need to know how the top (single) and bottom (triple)
magnets’ strengths compare. It would seem logical
that the triple magnet would be three times stronger
than the single, but in this part of the lab you’re go-
ing to find out for sure.
B Qualitative observations of the interaction of
two magnets
Play around with the two magnets and see how they
interact with each other. Can one attract the other?
Can one repel the other? Can they act on each other
simultaneously? Do they need to be touching in or-
der to do anything to each other? Can A act on B
while at the same time B does not act on A at all?
Can A pull B toward itself at the same time that
B pushes A away? When holding one of the heavier
magnets, it may be difficult to feel when there is any
push or pull on it; you may wish to have one person
hold the magnet with her eyes closed while the other
person moves the other magnet closer and farther.

C Measurement of interactions between two mag-

nets
Orient your magnet this way, as if it’s rolling toward the Once you have your data from parts A and B, you
compass from the north. With no magnet nearby, the are ready to form a hypothesis about the following
compass points to magnetic north (dashed arrow). The situation. Suppose we set up two balances as shown
magnet deflects the compass to a new direction. in the figure. The magnets are not touching. The
top magnet is hanging from a hook underneath the
pan, giving the same result as if it was on top of the
One way of measuring the strength of a magnet is pan. Make sure it is hanging under the center of the
to place the magnet to the north or south of the pan. You will want to make sure the magnets are
compass and see how much it deflects (twists) the pulling on each other, not pushing each other away,
needle of a compass. You need to test the magnets so that the top magnet will stay in one place.
at equal distances from the compass, which will pro-
duce two different angles.1 It’s also important to get
everything oriented properly, as in the figure.2
Make sure to take your data with the magnets far
enough from the compass that the deflection angle
is fairly small (say 5 to 30 ◦ ). If the magnet is close
enough to the compass to deflect it by a large an-
gle, then the ratio of the angles does not accurately
represent the ratio of the magnets’ strengths. After
all, just about any magnet is capable of deflecting
the compass in any direction if you bring it close
enough, but that doesn’t mean that all magnets are
equally strong.

1 There are two reasons why it wouldn’t make sense to find

different distances that produced the same angle. First, you

don’t know how the strengths of the effect falls off with dis-
tance; it’s not necessarily true, for instance, that the magnetic
field is half as strong at twice the distance. Second, the point
of this is to help you interpret part C, and in part C, the triple
The balances will not show the magnets’ true masses,
magnet’s distance from the single magnet is the same as the
single magnet’s distance from the first magnet. because the magnets are exerting forces on each other.
2 Laying the magnet flat on the table causes the compass The top balance will read a higher number than it
needle to try to tilt out of the horizontal plane, which it’s not would without any magnetic forces, and the bot-
designed to do. Turning it so that it faces the compass also
doesn’t work, because it makes the magnet’s magnetic field
tom balance will have a lower than normal reading.
lie along the same north-south line as the Earth’s, rather than The difference between each magnet’s true mass and
perpendicular to it. the reading on the balance gives a measure of how

7
strongly the magnet is being pushed or pulled by the
other magnet.
How do you think the amount of pushing or pulling
experienced by the two magnets will compare? In
other words, which reading will change more, or will
they change by the same amount? Write down a hy-
pothesis; you’ll test this hypothesis in part C of the
lab. If you think the forces will be unequal predict
their ratio.
Discuss with your instructor your results from parts
A and B, and your hypothesis about what will hap-
pen with the two balances.
Now set up the experiment described above with two
balances. Since we are interested in the changes in
the scale readings caused by the magnetic forces, you
will need to take a total of four scale readings: one
pair with the balances separated and one pair with
the magnets close together as shown in the figure
above.
When the balances are together and the magnetic
forces are acting, it is not possible to get both bal-
ances to reach equilibrium at the same time, because
sliding the weights on one balance can cause its mag-
net to move up or down, tipping the other balance.
Therefore, while you take a reading from one bal-
ance, you need to immobilize the other in the hori-
zontal position by taping its tip so it points exactly
at the zero mark.
You will also probably find that as you slide the
weights, the pointer swings suddenly to the oppo-
site side, but you can never get it to be stable in
the middle (zero) position. Try bringing the pointer
manually to the zero position and then releasing it.
If it swings up, you’re too low, and if it swings down,
you’re too high. Search for the dividing line between
the too-low region and the too-high region.
If the changes in the scale readings are very small
(say a few grams or less), you need to get the mag-
nets closer together. It should be possible to get the
scale readings to change by large amounts (up to 10
or 20 g).
Part C is the only part of the experiment where you
will be required to analyze random errors using the
techniques outlined in Appendices 2 and 3 at the
back of the lab manual. Think about how you can
get an estimate of the random errors in your mea-
surements. Do you need to do multiple measure-
ments? Discuss this with your instructor if you’re
uncertain.
Don’t take apart your setup until lab is over, and
8 Lab 1 Interactions
you’re completely done with your analysis — it’s no
fun to have to rebuild it from scratch because you
made a mistake!
D Measurement of interactions involving ob-
jects in contact
You’ll recall that Aristotle gave completely different
interpretations for situations where one object was
in contact with another, like the hand pushing the
ball, and situations involving objects not in contact
with each other, such as the rock falling down to
the earth. Your magnets were not in contact with
each other. Now suppose we try the situation shown
below, with one person’s hand exerting a force on the
other’s. All the forces involved are forces between
objects in contact, although the two people’s hands
cannot be in direct contact because the spring scales
have to be inserted to measure how strongly each
person is pulling. Suppose the two people do not
make any special arrangement in advance about how
hard to pull. How do you think the readings on the
two scales will compare? Write down a hypothesis,
and discuss it with your instructor before continuing.
Now carry out the measurement shown in the figure.
Self-Check
Do all your analysis in lab, including error analysis
for part C. Error analysis is discussed in appendices
2 and 3; get help from your instructor if necessary.
Analysis
In your writeup, present your results from all four
parts of the experiment, including error analysis for
part C.
done including interactions
Analysis
The most common mistake is to fail to address the
point of the lab. If you feel like you don’t understand
why you were doing any of this, then you were miss-
ing out on your educational experience! See the back
of the lab manual for the format of lab writeups.
Notes For Next Week
(1) Next week, when you turn in your writeup for
this lab, you also need to turn in a prelab writeup
for the next lab. The prelab questions are listed
at the end of the description of that lab in the lab
manual. Never start a lab without understanding
the answers to all the prelab questions; if you turn
in partial answers or answers you’re unsure of, dis-
cuss the questions with your instructor or with other
students to make sure you understand what’s going
on.
(2) You should exchange phone numbers with your
lab partners for general convenience throughout the
semester. You can also get each other’s e-mail ad-
dresses by logging in to Spotter and clicking on “e-
mail.”

Rules and Organization for

Professor Crowell’s Labs
Collection of raw data is work you share with your
lab partners. Once you’re done collecting data, you
need to do your own analysis. E.g., it is not okay for
two people to turn in the same calculations, or on a
lab requiring a graph for the whole group to make
one graph and turn in copies.
As explained in the syllabus, you’ll do some labs as
formal writeups, others as informal “check-off” labs.
The format of formal lab writeups is given in ap-
pendix 1 on page 161. The raw data section must
be contained in your bound lab notebook. Typically
people word-process the abstract section, and any
other sections that don’t include much math, and
stick the printout in the notebook to turn it in. The
justification and reasoning section will usually just
consist of hand-written calculations you do in your
lab notebook. You need two lab notebooks, because
on days when you turn one in, you need your other
one to take raw data in for the next lab. You may
find it convenient to leave one or both of your note-
books in the cupboard at your lab bench whenever
you don’t need to have them at home to work on;
this eliminates the problem of forgetting to bring
your notebook to school.
For a check-off lab, the main thing I’ll pay attention
to is your abstract. The rest of your work for a
check-off lab can be informal, and I may not ask to
see it unless I think there’s a problem after reading
your abstract.

9
2 Kinematics
Apparatus
computer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
track . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
dynamics cart . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
fan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
AA batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4/group
aluminum slugs . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
motion detector . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
protractor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
WD-40
Goal
Learn how to relate the motion of an object to its
position-versus-time graph.
Introduction
Analyzing motion is the most fundamental thing we
do in physics. The most versatile way of representing
motion is with a graph that has the object’s position
on the upright axis and time on the horizontal axis.
It takes some practice to be able to sketch and in-
terpret these graphs, but once you get used to them,
they become very intuitive.
Apparatus
The object whose motion you’ll study is a cart that
rolls on a track. You can either push the cart by
hand, start it moving with a shove, or clamp a fan on
top of it to make it speed up or slow down steadily.
To measure the cart’s motion, you’ll use a little sonar
gun that sends out clicks. When it hears the echo
from the cart, it figures out how far away the cart
was based on the time delay and the known speed of
sound. The sonar gun is connected to a computer,
which produces a position-versus-time graph.
10 Lab 2 Kinematics
Setup
Check that all four wheels on the cart will spin for
about 20-30 seconds if you flick them hard. If they
only spin for a few seconds, see if you can fix the
problem by spraying WD-40 on the bearings.
Set the cart on the track without the fan. Prop
the motion detector (sonar gun) at one end of the
track so that it is aimed slightly upward. This angle
is critical — measure 86 ◦ above horizontal with the
protractor, and tape it to the backrest.
With the computer turned off, plug the motion de-
tector into the PORT2 plug on the interface box.
Start up the computer. For compactness, I’ll use no-
tation like this to describe the computer commands:
Start>Programs>Vernier Software>Logger Pro 2
This is the command to start the computer soft-
ware running. “Start” means to click on the start
menu at the bottom left corner of the screen, “Pro-
grams” means to select that from the menu, and so
on. There are two different versions of the software
installed; use version 2. (Logger Pro 3 doesn’t work
with the interface boxes we have.)
Make sure that the interface box is plugged into
COM1 (the first COM port) at the back of the com-
puter, not COM2. If the computer presents you
with a dialog box saying “Set Up Interface,” choose
COM1.
Once the program is running, do File>Open, then
go into Probes and Sensors and then into Motion
Detector, and open the file of the same name. At
this point, you may get the following error message,
which you can ignore: “This file cannot run properly
with this hardware interface.”
You’ll get three graphs on the screen, but you only
want one, the x − t graph. Click on the x − t graph,
and then do View>Graph Layout>One Pane, and
the other two graphs will go away.
If you now click the button to tell it to collect data,
the motion detector should start clicking rapidly,
and it you move the cart back and forth you should
see a graph of its motion. Make sure it is able to
sense the cart’s motion correctly for distances from
50 cm to the full length of the track. If it doesn’t
work when the cart is at the far end of the track,
play with the angle of motion detector a little. If
you’re having other problems, you may find some
relevant debugging information in appendix 6, which
describes how to use the computer interface with a
different sensor.
Observations
In parts Athrough E,you don’t need to take detailed
numerical data — just sketch the graphs in your lab
notebook. All of your graphs will have garbage data
at the beginning and the end, and you need to make
sure you understand what’s what.
A Fast and slow motion
Moving the cart by hand, make a graph for slow
motion and another for fast motion. Make sure the
motion is steady, and don’t get confused by the parts
of the graph that come before and after your period
of steady pushing. Sketch the graphs and make sure
you understand them.
B Motion in two different directions
Now try comparing the graphs you get for the two
different directions of motion. Again, record what
they look like and figure out what you’re seeing.
C Reproducing a graph
Now see if you can produce a graph that looks like
this:
D Accelerating away from the sensor
Suppose the fan is mounted on the cart as shown
in the figure, so that if the cart is released from a
position close to the motion detector, it will begin
moving away from it. Predict what you think the
cart’s position-time graph will look like, and show
your prediction to your instructor before getting a
fan.
Before putting the batteries in the fan, make sure the
fan’s switch is off (to the right). Put the batteries in
and clamp the fan on the cart.
Set up the situation described above, and compare
the results with what you predicted.
E Slow or Rapid Acceleration
The aluminum slugs can be used to replace two of
the batteries so that the fan will exert about half as
much force. Discuss with your partners what you
think will happen if you repeat your previous run
with a weakened fan. Now try it.
F Changing the direction of motion
Change the fan back to full strength.
Now suppose instead of releasing the cart from rest
close to the motion detector, you started it moving
with a push toward the motion sensor, from the far
end of the track. It will of course slow down and
eventually come back. Discuss with your partners
what the position-time graph would look like. Now
try it.
G Rate of changing speed
The goal of this part of the lab is to determine
whether the speed of the cart in part F was changing
at a constant rate, i.e., by the same amount every
second.
Zoom in on the relevant part of your graph from part
F. To zoom in, either (a) draw a box with the mouse
and click on the magnifying glass icon, or (b) do
View>Graph Options>Axis Options to select ranges
of time and position values that you want. Print out
a big copy; choose landscape mode in the print dialog
box. (Note that if you take different data later, you
may need to fiddle with this again because you’ll be
zoomed in on the wrong part of the new graph.)
If that printer isn’t working, here’s what you need to
do instead. Do File>Export Data, and select “.txt”
for the type of the file. Use a text editor such as
WordPad to delete the header from the file. Save it
in your FC student directory, and also on a floppy
disk if you intend to work on it at home. Get into
OpenOffice or Excel, and open the file. Appendix 4
describes how to use OpenOffice. Whatever method
you use, make sure the whole group will end up with
copies.
Rather than trying to read distances from your graph’s
vertical axis in units of meters, and times from its
horizontal axis in units of seconds, the simplest thing
to do is simply to use a ruler to measure vertical
and horizontal distances on the graph, and deter-
mine the slopes from these; although the resulting
slopes won’t be in any standard units, that won’t
affect your conclusion.
11
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Make a prediction of the four graphs you’ll ob-
tain in parts A and B.

Self-Check
Do the analysis in lab.

Analysis
At one-second intervals, draw nice long tangent lines
on the curve from part G and determine their slope.
Some slopes will be negative, and some positive.
Summarize this series of changing speeds in a table.
Did the velocity increase by about the same amount
with every second?

12 Lab 2 Kinematics
13
3 Free Fall
Apparatus ideal case of θ=90 ◦ , which would be the same as free
fall. Galileo’s task would have been a lot simpler
two stations: if he’d had accurate enough devices for measuring
Behr free-fall column and weight time, because then he could have simply carried out
plumb bob measurements for objects falling vertically. That’s
spark generator (CENCO) what you’ll do today.
paper tape
switch for electromagnet

Goal
Find out whether it is ∆v/∆x or ∆v/∆t that is con-
stant for an object accelerating under the influence
of gravity.

Introduction
A fundamental and difficult problem in pre-Newton-
ian physics was the motion of falling bodies. Aristo-
tle had various incorrect but influential ideas on the
subject, including the assertions that heavier objects
fell faster than lighter ones and that the object only
sped up for a short while after it was dropped and
then continued on at a constant speed. Even among
Renaissance scientists who disagreed with Aristotle’s
claim that the object no longer sped up after a while,
there was a great deal of confusion about whether it
was ∆v/∆x or ∆v/∆t that was constant. It seems
obvious to modern physicists that they could not
both be constant, but it was not at all obvious to
authorities such as Domingo de Soto and Albert of
Saxony. Galileo started out thinking they were both
constant, then realized this was mathematically im-
possible, and finally determined from experiments
that it was ∆v/∆t, now called acceleration, that was
constant. A Setup
The main reason why the confusion persisted for two The apparatus consists of a 2-meter tall column with
thousand years was that the methods for measuring a paper tape running down it. A weight is held at the
time were inaccurate, and the time required for an top with an electromagnet and then released, falling
object to fall was very short. Galileo was able to right next to the paper tape. (An electromagnet
make settle the issue because he figured out how to is an artificial magnet that works when you put an
use a pendulum to measure time accurately, and also electric current through it, unlike a permanent mag-
came up with the idea of effectively slowing down the net, which does not require power.) A spark gener-
motion by studying objects rolling down an inclined ator is hooked up to the two vertical wires, and as
plane, rather than objects falling vertically. He then the weight falls, sparks cross the gap from the first
found how to extrapolate from the case of an object wire to the metal flange on the weight, then from
rolling down an inclined plane at an angle θ to the the flange to the other wire. Sparks are produced
only briefly, at regular intervals of 1/60 of a sec-

14 Lab 3 Free Fall

ond. On their way, the sparks go through the paper would be tedious to plot them all by hand.
tape, making dots on it that show the location of the
Determine whether your data are consistent with
weight at 1/60-s intervals.
constant ∆v/∆x or ∆v/∆t or neither.
First, unplug the spark generator so you don’t get
shocked while you’re getting things ready. Use the
switch made from a regular light switch to turn on Self-Check
the magnet at the top of the column, which operates
Appendix 4 discusses graphing. The graphing for
on 7 volts from the lab’s DC power circuits. Insert
this lab is time-consuming without a computer; since
the plumb bob, hanging from the magnet. Use the
we have a limited number of computers in lab, you
three screws on the feet of the column to level the
may want to go to one of the other campus computer
apparatus so the plumb bob’s string is parallel to
labs for this. Determine which quantity is constant.
the wire.
Replace the plumb bob with the weight. Pull fresh
tape up from the roll at the bottom, and get the Prelab
tape straight and centered on the wire.
The point of the prelab questions is to make sure
Plug in the spark generator, and put the function you understand what you’re doing, why you’re do-
knob on “line,” which means it will base its cycle ing it, and how to avoid some common mistakes. If
of sparks on the AC power from the wall, which you don’t know the answers, make sure to come to
switches directions once every sixtieth of a second. my office hours before lab and get help! Otherwise
The red LED should light up. From now on, do not you’re just setting yourself up for failure in lab.
press the thumb switch to activate the sparks unless
P1 Consider the quantities x, ∆x, t, and ∆t, which
you are sure nobody is near the vertical wires. Try
are measured more or less directly in this lab. Which
it out, and see if you get a spot at the top of the
of these would have a single value throughout the
tape, where the weight currently is.
whole motion of the falling weight?
B Observations P2 Explain, based on the meaning of the symbols
Hold down the thumb switch to make the sparks ∆ and /, why ∆v/∆x and ∆v/∆t have to be nota-
start, flip the switch to release the weight, and wait tions for numbers, not descriptions of graphs. How
until the weight has fallen in the cup at the bottom would these two numbers relate to the two graphs?
before releasing the thumb switch. You want a nice
P3 Suppose that, once you have data, ∆v/∆x turns
straight line of dots on the tape, going all the way
out to be constant, and ∆v/∆t isn’t. Explain how
from the top to the bottom — you may have to make
you would be able to tell this from your graphs.
adjustments and try a few times before getting a
good tape. Take your tape off, and measure the
locations of the dots accurately with a two-meter
stick.

Analysis
Since the sparks start before you release the electro-
magnet, the first dot at the very top of the tape will
give the starting position of the weight.
If you consider any adjacent pair of dots (avoiding
the top and bottom ones), then measuring the dis-
tance ∆x between them allows you to calculate an
approximation to the speed of the weight, which you
can think of as being its speed at the point half-way
between the two dots.
Make one plot of speed versus time and another of
speed versus distance, preferably using a computer,
since you will have about thirty data points, and it

15
4 The Earth’s Gravitational Field
Note to the lab technician: The computers to use
are the ones with labels on their sides saying Lassie,
Fang, and Buck.
Apparatus
(two stations):
vertical plank with electromagnets
steel balls (2/station)
Linux computers with Audacity installed (in 416T)
Goal
Make a high-precision measurement of the strength
of the Earth’s gravitational field, g, in Fullerton.
Introduction
When objects fall, and all forces other than grav-
ity are negligible, we observe that the acceleration
is the same, regardless of the object’s mass, shape,
density, or other properties. However, the acceler-
ation does depend a little bit where on the earth
we do the experiment, and even bigger variations in
acceleration can be observed by, e.g., going to the
moon. Thus, this acceleration can be considered as
a property of space itself, and we can refer to it as
the gravitational field in that region of space. Just
as you would use a magnetic compass to find out
about the magnetic field in the classroom, you can
use dropping masses to find out about the gravita-
tional field. In this experiment, you’ll measure the
gravitational field, g, in the classroom to sufficiently
high precision that, if everybody does a good job and
we pool and average everyone’s data to reduce ran-
dom errors, we should be able to get a value that is
measurably different from the generic world-average
value you would find in a textbook.
A Measuring g precisely
You will measure g, the acceleration of an object in
free fall, using electronic timing techniques. The idea
of the method is that you’ll have two steel balls hang-
ing underneath electromagnets at different heights.
You’ll simultaneously turn off the two magnets us-
ing the same switch, causing the balls to drop at
the same moment. The ball dropped from the lower
16 Lab 4 The Earth’s Gravitational Field
height (h1 ) takes a smaller time (∆t1 ) to reach the
floor, and the ball released from the greater height
(h2 ) takes a longer time (∆t2 ). The time intervals
involved are short enough that due to the limita-
tions of your reflexes it is impossible to make good
enough measurements with stopwatch. Instead, you
will record the sounds of the two balls’ impacts on
the floor using the computer. The computer shows
a graph in which the x axis is time and the y axis
shows the vibration of the sound wave hitting the
microphone. You can measure the time between the
two visible “blips” on the screen. You will measure
three things: h1 , h2 , and the time interval ∆t2 − ∆t1
between the impact of the second ball and the first.
From these data, with a little algebra, you can find
g.
The experiment would have been easier to analyze if
we could simply drop a single ball and measure the
time from when it was released to when it hit the
floor. But since our timing technique is based on
sound, and no sound is produced when the balls are
released, we need to have two balls. If h1 , the height
of the lower ball, could be made very small, then it
would hit the floor at essentially the same moment
the two balls were released (∆t1 would equal 0), and
∆t2 −∆t1 would be essentially the same as ∆t2 . But
we can’t make h1 too small or the sound would not
be loud enough to detect on the computer.
B Using the computer software
There are three Linux computers in 416T that have
the right software and hardware. They’re labeled
Lassie, Fang, and Buck. (On Fang, only sound in-
puyt works, not output.) First let’s see how to record
yourself on the computer saying “hello.” Use the
xmix or xmixer program to set the record and mic
levels all the way up. Start up the sound recording
program, called Audacity. Set the record level on
high, using the control marked − . . . + next to the
microphone icon. Record your sound.
Before you get down to serious science, you may en-
joy listening to your own voice reversed in time. A
fun diversion is to write a sentence down backwards,
read it out loud, and then electronically reverse it
so it’s forward again. It sounds sort of like someone
with a thick Hungarian accent.
To find out how long a sound is, you can use the
curson and click to find the time corresponding to
 To drop the balls, you should break the circuit
by pulling one of the wires out of the plug on
the power supply. Don’t use the switch; if you
use the switch, it takes some time for the mag-
netic field to decay, and the balls stick for a
while before dropping.
Two thumps, as recorded on the computer through the
Audacity will let you keep on making new record-
microphone.
ings, stacking the graphs vertically. However,
if you do this you will introduce significant tim-
a particular point in the graph. Sometimes you are ing errors. The reason is probably that Audac-
not sure which wiggles in the visual representation of ity is designed for use in multitrack recording
the sound correspond to which parts of the recorded of music, so it tries to play back the previously
sound. To find out, you can select part of the sound recorded tracks while recording the new one,
and listen to only that part. and on cheap sound hardware this causes little
timing glitches.
C Finding the interval between two sounds
When you record the sound of the two consecutive
impacts of the balls, they will look like vertical spikes Analysis
on the screen. You can practice using hand claps.
Extract a value of g from your data.
To accurately find the time when one of the sounds
Derive error bars on your result, using the techniques
started, first zoom in on it until it’s like you’re seeing
in appendices 2 and 3.
it under a powerful microscope. Click on the onset
of the sound, and read the time from the scale at the
top of the window. Self-Check
You may get this error message: ”Error while open-
Extract the value of g, with error bars. Read Ap-
ing sound device. Please check the input device set-
pendix 3 for information on how to do error analysis
tings and the project sample rate.” Quit the program
with propagation of errors; get help from your in-
and restart it.
structor if necessary.
Some of the computers have very low gain on their
microphones, so it may be difficult to see the thumps.
To work around this, do the following after you’ve Prelab
recorded a sound: type control-A to select the whole
The point of the prelab questions is to make sure
recording, and then do Effect>Amplify; the default
you understand what you’re doing, why you’re do-
is to amplify the sound by the maximum possible
ing it, and how to avoid some common mistakes. If
amount,which is what you want.
you don’t know the answers, make sure to come to
You should make a series of measurements, and make my office hours before lab and get help! Otherwise
sure they agree at the level of about 10−4 s; if they you’re just setting yourself up for failure in lab.
don’t, there’s something wrong with your technique.
P1 If your instructor has assigned homework prob-
Also, you should check that your result for g makes
lem 27 from ch. 3 of Newtonian Physics,don’t bother
sense.
turning in another copy of your work for this prelab
Here are some common problems that cause incon- question. Derive an equation for g in terms of the
sistent or wrong results: quantities you’ll measure, which are h1 , h2 , and the
time interval ∆t2 − ∆t1 . The point of the lab is to
The balls are brushing against the electrical measure g, so don’t just say “well of course g is 9.8
wires as they fall. m/s2 .” (You should check your equation by using
the answer checker for the homework problem.)
You’re misidentifying the thumps.
The surface the balls are dropping onto has
dents in it.
You’re not positioning the balls in on the same
spot on the magnets every time.

17
5 Newton’s Second Law
Apparatus of masses, but keep the total amount of mass con-
stant and just divide it differently between the two
pulley holders. Remember to take the masses of the holders
spirit level themselves into account. Make sure to perform your
string measurements with the longest possible distance of
weight holders, not tied to string travel, because you cannot use a stopwatch to get an
two-meter stick accurate measurement of very short time intervals.
slotted weights The best results are obtained with combinations of
stopwatch weights that give times of about 4 to 20 seconds.
foam rubber cushions Also, make sure that the masses are at least a few
hundred grams or so on each side.

Goal Self-Check
Find the acceleration of unequal weights hanging Compare theoretical and experimental values of ac-
from a pulley. celeration for one of your mass combinations. Check
whether they come out fairly consistent.

Observations
Analysis
Use your measured times and distances to find the
actual acceleration, and make a graph of this versus
M − m. Show these experimentally determined ac-
celerations as small circles. Overlaid on the same
graph, show the theoretical equation as a line or
curve.

Set up unequal masses on the two sides of the pulley,
and determine the resulting acceleration by measur-
ing how long it takes for the masses to move a cer-
tain distance. Use the spirit level to make the pulley
vertical; otherwise you get extra friction. Use rela-
tively large masses (typically half a kg or a kg each
side) so that friction is not such a big force in com-
parison to the other forces, and the inertia of the
pulley is negligible compared to the inertia of the
hanging masses. Do several different combinations
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Criticize the following reasoning: The weight
fell 1.0 m in 1 s, so v = 1 m/s, and a = v/t = 1 m/s2 .
P2 Since that won’t work, plan how you really will
determine your experimental accelerations based on
your measured distance and times.

18 Lab 5 Newton’s Second Law

19
6 Air Friction
Apparatus push harder against the road.

coffee filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10/group Your goal in this lab is to find a proportionality re-
stopwatch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group lating the force of air friction to the velocity at which
computer with sonar sensor the air rushes over the object. For instance, you may
wire cages to protect sensors find the rule
F ∝v ,
which is a shorthand for

Goal F = (some number)(v) .

Determine how the force of air friction depends on

The numerical value of “some number” is not very
the velocity of a moving object.
interesting, because we would expect it to be dif-
As an alternative, you may create your own tech- ferent for different objects, which is why you would
nique for doing the same sort of experiment for fric- write your result as F ∝ v. This proportionality
tion between two surfaces wetted with a liquid such would tell you for instance that anytime the speed
as water, vegetable oil, or machine oil — the result was doubled, the result would be twice as much air
might be more interesting, since it is not to be found friction.
in textbooks. If you are interested in doing this, dis-
Suppose instead you find that doubling the speed
cuss it in advance with your instructor.
makes the force eight times greater, multiplying the
speed by 10 makes the force 1000 times greater, and
so on. In each case, the force is being multiplied
Introduction by the third power of the increase in the speed, i.e.,
Friction between solid objects occurs all the time in F ∝ v3 .
our daily lives. The frictional force exerted by the air
on a solid object is not as often evident, but it is re-
sponsible for the wind blowing our hair, for the slow Observations
dropping of a feather, and for our cars’ poorer gas
There are two possible methods for measuring the
mileage at freeway speeds compared to more mod-
terminal velocity of the filter.
erate speeds.
The first method is shown in the figure below. We
The latter effect suggests that air friction might in-
use coffee filters because they don’t tumble or sway
crease with speed, unlike solid-solid friction, which is
very much as they fall, and because they allow us to
nearly independent of speed. By Newton’s first law,
easily change the mass of our falling object by nest-
a car or a jet plane cruising at constant speed must
ing more coffee filters inside the bottom one, with-
have zero total force on it, so if the air friction force
out changing its aerodynamic properties. The filters
gets stronger with speed, that would explain why
will start speeding up when you release them near
a greater forward-pushing force would be needed to
the ceiling, but as they speed up, the upward force
travel at high speeds. For instance, a car traveling
of air friction on them increases, until they reach a
at low speed might have a -10 kN air friction force
speed at which the total force on them is zero. Once
pushing backward on it, so in order to have zero to-
at this speed, they obey Newton’s first law and con-
tal force on it the road must be making a forward
tinue at constant speed. If the number of coffee fil-
force of +10 kN. At a higher speed, air friction might
ters is small, they will have reached their maximum
increase to -30 kN, so the road would need to make a
speed within the first half a meter or so. By the
forward force of +30 kN. The car convinces the road
time they are even with the edge of the lab bench,
to make the stronger force by pushing backward on
they are moving at essentially their full speed. You
the road more strongly: by Newton’s third law, the
can then use the stopwatch to determine how long it
car’s force on the road and the road’s force on the
takes them to cover the distance to the floor, which
car must be equal in magnitude and opposite in di-
will allow you to find their speed. During this final
rection. The car burns more gas because it must
part of the fall, you know the upward force of air

20 Lab 6 Air Friction

friction must be as great as the downward force of
gravity, so you can determine what it was.
A different technique is to drop the filters onto a
sonar sensor of the type used in lab 2. You can put
the sensor on the floor facing up, and put the wire
cage over it to keep it from getting damaged by being
stepped on inadvertently.
For a long time, I had my students do the lab using
only the first method, but now I’m experimenting
with the second method. A couple of advantages of
the second method are that (1) it doesn’t depend on
human reflexes, and (2) it gives you a real-time pic-
ture of the motion, so it’s easier to tell whether the
filters are actually reaching terminal velocity. The
latter is an important issue, because it gives you a
better chance of being able to take data over a wide
range of values for F and v, but without running
into problems with cases where the filters don’t re-
ally reach terminal veloicity. On the other hand, the
computer method has some practical problems, such
as the tendency of the filters to drift sideways instead
of heading straight down onto the sensor. This is an
opportunity for you to do something like what real
scientists do: use your ingenuity and try different
things to see what works best!
Note that if the coffee filters get too flattened out,
they’ll flutter, giving lousy results.
Take data with stacks of various numbers of coffee
filters. You will get the most clearcut determination
of the power law relationship if your data cover the
largest possible range of values. It’s a good idea to
take some data with a large number of filters, drop-
ping them from the balcony outside so they have
time to get up to their final speed. This is also the
only way you can tell for sure whether you’re tak-
ing data at terminal velocity: the results at the two
different heights (inside and outside) should be con-
sistent.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Suppose you tried to do this lab with stacks of
coins instead of coffee filters. Assuming you had a
sufficiently accurate timing device, would it work?
P2 Criticize the following statement:
“We found that bigger velocities gave bigger air drag
forces, which demonstrates the proportionality F ∝
v.”
P3 Criticize the following statement:
“We found F ∝ v 7 , which shows that you need more
force to make things go faster.”
Analysis
Use your raw data to compile a list of F and v values.
Use the methods explained in Appendix 5 to see if
you can find a power-law relationship between F and
v. This will require fitting a line to a set of data, as
explained in appendix 4. Both fitting a line to data
and finding power laws are techniques you will use
several more times in this course, so it is worth your
while to get help now if necessary in order to get
confident with them.
21
7 Vector Addition of Forces
Apparatus
force table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
spirit level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
weights
string

Goal
Observations
Test whether the vector sum of the forces acting on
an object at rest is equal to zero. The apparatus consists of a small circular table, with
a small metal ring held in the middle by the tension
in four strings. Each string goes over a pulley at the
Introduction edge of the table, so that a weight can be hung on it
to control the tension. The angles can be recorded
Modern physics claims that when a bridge, an earth-
either graphically, by sliding a piece of paper un-
quake fault, or an oak tree doesn’t move, it is be-
derneath, or by reading angles numerically off of an
cause the forces acting on it, which combine accord-
angular scale around the circumference of the table.
ing to vector addition, add up to zero. Although
this may seem like a reasonable statement, it was Use the spirit level to level the table completely us-
far from obvious to premodern scientists. Aristotle, ing the screws on the feet. Set up four strings with
for instance, said that it was the nature of each of weights, using the small pin to hold the ring in place.
the four elements, earth, fire, water and air, to re- Adjust the angles or the amounts of weight or both,
turn to its natural location. Rain would fall from until the ring is in equilibrium without the pin, and
the sky because it was trying to return to its natu- is positioned right over the center of the table. Avoid
ral location in the lakes and oceans, and once it got a symmetric arrangement of the strings (e.g., don’t
to its natural location it would stop moving because space them all 90 degrees apart), and don’t make any
that was its nature. forces collinear with each other. The ring is an ex-
tended object, so in order to treat it mathematically
When a modern scientist considers a book resting on
as a pointlike object you should make sure that all
a table, she says that it holds still because the force
the strings are lined up with the center of the ring,
of gravity pulling the book down is exactly canceled
as shown in the figure.
by the normal force of the table pushing up on the
book. Aristotle would have denied that this was pos-
sible, because he believed that at any one moment an
object could have only one of two mutually exclusive
types of motion: natural motion (the tendency of the
book to fall to the ground, and resume its natural
place), and forced motion (the ability of another ob-
ject, such as the table, to move the book). According
to his theory, there could be nothing like the addi-
tion of forces, because the object being acted on was
only capable of “following orders” from one source at
a time. The incorrect Aristotelian point of view has Because of friction, it is possible to change any one
great intuitive appeal, and beginning physics stu- of the weights slightly without causing the ring to
dents tend to make Aristotelian statements such as, move. This is a potential source of systematic er-
“The table’s force overcomes the force of gravity,” rors, but you can eliminate the error completely by
as if the forces were having a contest, in which the the following method. Find out how much you can
victor annihilated the loser. increase or decrease each weight without causing the
ring to move. Within the range of values that don’t

22 Lab 7 Vector Addition of Forces

cause slipping, use the center of the range as your the magnitude of the sum of the forces is small com-
best value; with this amount of weight, there is no pared to the magnitudes of the individual forces.
friction at all in the pulley. The point here is not to
redo the entire experiment with a completely differ-
ent combination of weights — that would not tell Analysis
you anything about friction as a source of error,
Calculate the magnitude of vector sum of the forces
since even if there was no friction at all, it would be
on the ring, first graphically and then analytically.
possible for example to double all the weights and
Make sure the two methods give the same result. If
get an equilibrium. Once you’ve set each weight to
they do not, try measuring the x and y components
its friction-free value, leave it that way; by the time
off of your drawing and comparing them with the x
you’re done, you will have eliminated friction from
and y components you calculated analytically.
all four pulleys.
Estimate the possible random error in your final
sum.
Prelab Are your results consistent with theory, taking into
The point of the prelab questions is to make sure account the random errors involved?
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 The weights go on weight holders that hang
from the string, and the weight holders are each 50
g. Criticize the following reasoning: “We don’t need
to count the mass of the weight holders, because it’s
the same on all four strings, so it cancels out.”
P2 Describe a typical scale that you might use for
drawing force vectors on a piece of paper, e.g., how
long might you choose to make a 1-N force? Assume
your masses are from 500 to 1500 grams.
P3 Graphically calculate the vector sums of the
two pairs of vectors shown below. As a check on
your results, you should find that the magnitudes of
the two sums are equal. Lab 8 has been deleted.

Self-Check
Do both a graphical calculation and an analytic cal-
culation in lab, without error analysis. Make sure
they give the same result. Do a rough check that

23
9 Acceleration In Two Dimensions
Apparatus
air track (small)
cart
photogate (PASCO) (under lab benches in rm. 418)
computer
air blowers
power strips for switching CENCO blowers on and
off
vernier calipers
wood blocks
Goal
Test whether the acceleration of gravity acts like a
vector.
Introduction
As noted in lab 2, one of the tricky techniques Galileo
had to come up with to study acceleration was to
use objects rolling down an inclined plane rather
than falling straight down. That slowed things down
enough so that he could measure the time intervals
using a pendulum clock. Even though you were able,
in lab 4, to use modern electronic timing techniques
to measure the short times involved in a vertical fall,
there is still some intrinsic interest in the idea of
motion on an inclined plane. The reason it’s worth
studying is that it reveals the vector nature of accel-
eration.
Vectors rule the universe. Entomologists say that
God must have had an inordinate fondness for bee-
tles, because there are so many species of them.
Well, God must also have had a special place in her
heart for vectors, because practically every natural
phenomenon she invented is a vector: gravitational
acceleration, electric fields, nuclear forces, magnetic
fields, all the things that tie our universe together
are vectors.
Setup
The idea of the lab is that if acceleration really acts
like a vector, then the cart’s acceleration should equal
the component of the earth’s gravitational accelera-
tion vector that is parallel to the track, because the
24 Lab 9 Acceleration In Two Dimensions
cart is only free to accelerate in the direction along
the track. There is almost no friction, since the cart
rides on a cushion of air coming through holes in the
track.
The speed of the cart at any given point can be mea-
sured as follows. The photogate consists of a light
and a sensor on opposite sides of the track. When
the cart passes by, the cardboard vane on top blocks
the light momentarily, keeping light from getting to
the sensor. The computer detects the electrical sig-
nal from the sensor, and records the amount of time,
tb , for which the photogate was blocked. Given tb ,
you can determine the approximate speed that cart
had when it passed through the photogate. The use
of the computer software is explained in Appendix 6;
of the three modes described there, you want to use
the software in the mode in which it measures the
time interval over which the photogate was blocked.
Plug the photogate into the DG1 plug on the inter-
face box.
Observations
The basic idea is to release the cart at a distance x
away from the photogate. The cart accelerates, and
you can determine its approximate speed, v, when it
passes through the photogate. (See prelab question
P1. Make sure to use vernier calipers to measure the
width of the vane, w.) From v and x, you can find
the acceleration. You will take data with the track
tilted at several different angles, to see whether the
cart’s acceleration always equals the component of g
parallel to the track.
You can level the track to start with by adjusting
the screws until the cart will sit on the track without
accelerating in either direction.
The distance x can be measured from the starting
position of the cart to half-way between the point
where it first blocks the photogate and the point
where it unblocks the photogate. You can determine
where these positions are by sliding the cart into the
photogate and watching the red LED on the top of
the photogate, which lights up when it is blocked.
Hints:
Keep in mind that if the cart rebounds at the
bottom of the track and comes back up through
the gate, you will get a second, bogus time
reading.
Note that you have no way to measure accu-
rately to the total amount of time over which
the cart picked up speed (which would be sev-
eral seconds) — what you measure is the very
short time required for the cart to pass through
the photogate.
If you’re using one of the gray air pumps, which
has a knob to adjust the flow, make sure it’s on
the highest speed, or the cart will drag on the
track, giving bogus data. It’s easy to mess up
this adjustment, so get the knob set correctly
for once and for all, and then never touch it
again. To turn the pump on and off, plug the
blower in to its own power strip, and use the
switch on the power strip.
Release the cart by hand after starting up the
air pump. If you leave the cart on the track
and then turn on the pump, there will be a
period of time when the pump is first starting
up, and the cart will drag.
The variable x actually changes a little when
you change θ, so don’t just assume it’s always
the same.
Once you have data at your first angle, check
whether theory and experiment agree reason-
ably well.
You’ll use the photogates again in lab 10, so make
sure you understand the technique thoroughly, and
take notes on it so you’ll remember how it’s done.
Self-Check
Find the theoretical and experimental accelerations
for one of your angles, and see if they are roughly
consistent.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Skip this question if the corresponding home-
work problem from Newtonian Physics has already
been assigned. (a) If w is the width of the vane, and
tb is defined as suggested above, what is the speed of
the cart when it passes through the photogate? (b)
Based on v and x, how can you find a?
P2 Should x be measured horizontally, or along
the slope of the track?
P3 It is not possible to measure θ accurately with
a protractor. How can θ be determined based on
the distance between the feet of the air track and
the height of the wood block?
P4 Explain why the following method for finding
the cart’s acceleration is incorrect. “The time I got
off the computer was 0.0237 s. My vane was 2.2 cm
wide, so v = 2.2 cm/.0237 s = 93 cm/s. That means
the acceleration was a = ∆v/∆t = (vf − vi )/∆t =
(vf − 0)/∆t = vf /∆t, or 93 cm/s divided by .0237
s, which gives 3900 cm/s2 .”
Analysis
Extract the acceleration for each angle at which you
took data. Make a graph with θ on the x axis and
acceleration on the y axis. Show your measured ac-
celerations as points, and the theoretically expected
dependence of a on θ as a smooth curve.
Error analysis is not required for this lab, because
the random errors are small compared to systematic
errors such as the imperfect leveling of the track,
friction, warping of the track, and the measurement
of w.
25
10 Conservation Laws
Apparatus
Part A: vacuum pump (Lapine) . . . . . . . . . . . . . . . . . 1
electronic balance (large capacity) . . . . . . . . . . . . . . . 1
plastic-coated flask . . . . . . . . . . . . . . . . . . . . . . . 1/group
Part B: beaker . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
propyl alcohol 200 mL/group
canola oil 200 mL/group
funnels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
100-mL volumetric flask . . . . . . . . . . . . . . . . . . 1/group
rubber stopper, fitting in
volumetric flask . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
1-ml pipette and bulb . . . . . . . . . . . . . . . . . . . . 1/group
magnetic stirrer . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
triple-beam balance . . . . . . . . . . . . . . . . . . . . . . .1/group
Goal
People believe that objects cannot be made to disap-
pear or appear. If you start with a certain amount
of matter, there is no way to increase or decrease
that amount. This type of rule is called a conser-
vation law in physics, and this specific law states
that the amount of matter is conserved, i.e., must
stay the same. In order to make this law scientifi-
cally useful, we must define more carefully how the
“amount” of a substance is to be defined and mea-
sured numerically. Specifically, there are two issues
that scientifically untrained people would probably
not agree on:
Should air count as matter? If it has weight,
then it probably should count. In this lab, you
will find out if air has weight, and, if so, mea-
sure its density.
Should the amount of a substance be defined in
terms of volume, or is mass more appropriate?
In this lab, you will determine whether mass
and/or volume is conserved when water and
alcohol are mixed.
Introduction
Styles in physics come and go, and once-hallowed
principles get modified as more accurate data come
along, but some of the most durable features of the
science are its conservation laws. A conservation law
26 Lab 10 Conservation Laws
is a statement that something always remains con-
stant when you add it all up. Most people have a
general intuitive idea that the amount of a substance
is conserved. That objects do not simply appear
or disappear is a conceptual achievement of babies
around the age of 9-12 months. Beginning at this
age, they will for instance try to retrieve a toy that
they have seen being placed under a blanket, rather
than just assuming that it no longer exists. Con-
servation laws in physics have the following general
features:
Physicists trying to find new conservation laws
will try to find a measurable, numerical quan-
tity, so that they can check quantitatively whether
it is conserved. One needs an operational def-
inition of the quantity, meaning a definition
that spells out the operations required to mea-
sure it.
Conservation laws are only true for closed sys-
tems. For instance, the amount of water in a
bottle will remain constant as long as no wa-
ter is poured in or out. But if water can get in
or out, we say that the bottle is not a closed
system, and conservation of matter cannot be
applied to it.
The quantity should be additive. For instance,
the amount of energy contained in two gallons
of gasoline is twice as much as the amount of
energy contained in one gallon; energy is addi-
tive. An example of a non-additive quantity is
temperature. Two cups of coffee do not have
twice as high a temperature as one cup.
Conservation laws always refer to the total amount
of the quantity when you add it all up. If you
add it all up at one point in time, and then
come back at a later point in time and add it
all up, it will be the same.
How can we pin down more accurately the concept
of the “amount of a substance”? Should a gallon
of shaving cream be considered “more substantial”
than a brick? At least two possible quantities come
to mind: mass and volume. Is either conserved?
Both? Neither? To find out, we will have to make
measurements.
We can measure mass by the “see-saw method” —
when two children are sitting on the opposite sides
of a see-saw, the less massive one has to move far-
ther out from the fulcrum to make it balance. If we
enslave some particular child as our permanent mass
standard, then any other child’s mass can be mea-
sured by balancing her on the other side and mea-
suring her distance from the fulcrum. A more prac-
tical version of the same basic principle that does
not involve human rights violations is the familiar
pan balance with sliding weights.
Volume is not necessarily so easy to measure. For
instance, shaving cream is mostly air, so should we
find a way to measure just the volume of the bub-
bly film itself? Precise measurements of volume can
most easily be done with liquids and gases, which
conform to a vessel in which they are placed.
Should a gas, such as air, be counted as having any
substance at all? Empedocles of Acragas (born ca.
492 BC) was the originator of the doctrine that all
material substances are composed of mixtures of four
elements: earth, fire, water and air. The idea seems
amusingly naive now that we know about the chem-
ical elements and the periodic table, but it was ac-
cepted in Europe for two thousand years, and the
inclusion of air as a material substance was actu-
ally a nontrivial concept. Air, after all, was invis-
ible, seemed weightless, and had no definite shape.
Empedocles decided air was a form of matter based
on experimental evidence: air could be trapped un-
der water in an inverted cup, and bubbles would be
released if the cup was tilted. It is interesting to
note that in China around 300 BC, Zou Yan came
up with a similar theory, and his five elements did
not include air.
Does air have weight? Most people would probably
say no, since they do not feel any physical sensation
of the atmosphere pushing down on them. A delicate
house of cards remains standing, and is not crushed
to the floor by the weight of the atmosphere.
Compare that to the experience of a dolphin, though.
A dolphin might contemplate a tasty herring sus-
pended in front of it and conjecture that water had
no weight, because the herring did not involuntarily
shoot down to the sea floor because of the weight of
the water overhead. Water does have weight, how-
ever, which a sufficiently skeptical dolphin physicist
might be able to prove with a simple experiment.
One could weigh a 1-liter metal box full of water and
then replace the water with air and weigh it again.
The difference in weight would be the difference in
weight between 1 liter of water of and 1 liter of air.
Since air is much less dense than water, this would
approximately equal the weight of 1 liter of water.
Our situation is similar to the dolphin’s, as was first
appreciated by Torricelli, whose experiments led him
to conclude that “we live immersed at the bottom
of a sea of...air.” A human physicist, living her life
immersed in air, could do a similar experiment to
find out whether air has weight. She could weigh a
container full of air, then pump all the air out and
weigh it again. When all the matter in a container
has been removed, including the air, we say that
there is a vacuum in the container. In reality, a
perfect vacuum is very difficult to create. A small
fraction of the air is likely to remain in the container
even after it has been pumped on with a vacuum
pump. The amount of remaining air will depend
on how good the pump is and on the rate at which
air leaks back in to the container through holes or
cracks.
Cautions
Please do not break the glassware! The vacuum
flasks and volumetric flasks are expensive.
The alcohol you will be using in this lab is chemically
different from the alcohol in alcoholic beverages. It
is poisonous, and can cause blindness or death if you
drink it. It is not hazardous as long as you do not
drink it.
Observations
A Density of air
You can remove the air from the flask by attach-
ing the vacuum pump to the vacuum flask with the
rubber and glass tubing, then turning on the pump.
You can use the scale to determine how much mass
was lost when the air was evacuated.
Make any other observations you need in order to
find out the density of air and to estimate error bars
for your result.
B Is volume and/or mass conserved when two
fluids are mixed?
The idea here is to find out whether volume and/or
mass is conserved when water and alcohol are mixed.
The obvious way to attempt this would be to mea-
sure the volume and mass of a sample of water, the
volume and mass of a sample of alcohol, and their
volume and mass when mixed. There are two prob-
lems with the obvious method: (1) when you pour
one of the liquids into the other, droplets of liquid
will be left inside the original vessel; and (2) the
27
most accurate way to measure the volume of a liq-
uid is with a volumetric flask, which only allows one
specific, calibrated volume to be measured.
Here’s a way to get around those problems. Put the
magnetic stirrer inside the flask. Pour water through
a funnel into a volumetric flask, filling it less than
half-way. (Do not use the pipette to transfer the
water.) A common mistake is to fill the flask more
than half-way. Now pour a thin layer of cooking
oil on top. Cooking oil does not mix with water,
so it forms a layer on top of the water. (Set aside
one funnel that you will use only for the oil, since
the oil tends to form a film on the sides.) Finally,
gently pour the alcohol on top. Alcohol does not mix
with cooking oil either, so it forms a third layer. By
making the alcohol come exactly up to the mark on
the calibrated flask, you can make the total volume
very accurately equal to 100 mL. In practice, it is
hard to avoid putting in too much alcohol through
the funnel, so if necessary you can take some back
out with the pipette.
If you put the whole thing on the balance now, you
know both the volume (100 mL) and the mass of
the whole thing when the alcohol and water have
been kept separate. Now, mix everything up with
the magnetic stirrer. The water and alcohol form a
mixture. You can now test whether the volume or
mass has changed.
If the mixture does not turn out to have a volume
that looks like exactly 100 mL, you can use the fol-
lowing tricks to measure accurately the excess or
deficit with respect to 100 mL. If it is less than 100
28 Lab 10 Conservation Laws
mL, weigh the flask, pipette in enough water to bring
it up to 100 mL, weigh it again, and then figure out
what mass and volume of water you added based on
the change in mass. If it is more than 100 mL, weigh
the flask, pipette out enough of the mixture to bring
the volume down to 100 mL, weigh it again, and
make a similar calculation using the change in mass
and the density of the oil. If you need to pipette out
some oil, make sure to wash and rinse the pipette
thoroughly afterwards.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Give an example of two things having the same
mass and different densities.
P2 Give an example of two things having the same
density and different masses.
P3 Why can the density of water be given in a
book as a standard value under conditions of stan-
dard temperature and pressure, while the mass of
water cannot?
P4 What would your raw data in part A be like if
air had no weight? What would they be like if air
did have weight?
P5 Referring to appendix 2, plan how you will es-
timate your random errors.
P6 In part B, pick either mass or volume, and de-
scribe what your observations would be like if that
quantity was not conserved.
Self-Check
Do a quick analysis of both parts without error anal-
ysis. Plan how you will do your error analysis.
Analysis
A. If your results show that air has weight, determine
the (nonzero) density of air, with an estimate of your
random errors.
B. Decide whether volume and/or mass is conserved
when alcohol and water are mixed, taking into ac-
count your estimates of random errors.
29
11 Conservation of Energy
Apparatus
air track
cart
springs (steel, 1.5 cm diameter)
photogate (PASCO)
computer
stopwatches
air blowers
alligator clips
spring scales
vernier calipers
power strips for CENCO blowers
string
Goal
Test conservation of energy for an object oscillating
around an equilibrium position.
This could be a vibration of the sun, a water balloon, or
a nucleus.
Introduction
One of the most impressive aspects of the physical
world is the apparent permanence of so many of its
parts. Objects such as the sun or rocks on earth
have remained unchanged for billions of years, so it
might seem that they are in perfect equilibrium, with
zero net force on each part of the whole. In reality,
the atoms in a rock do not sit perfectly still at an
equilibrium point — they are constantly in vibration
about their equilibrium positions. The unchanging
oblate shape of the sun is also an illusion. The sun
is continually vibrating like a bell or a jiggling water
balloon, as shown in the (exaggerated) figure. The
nuclei of atoms also jiggle spontaneously like little
water balloons. The fact that these types of motion
continue indefinitely without dying out or building
up relates to conservation of energy, which forbids
them to get bigger or smaller without transferring
energy in or out.
Our model of this type of oscillation about equilib-
rium will be the motion of a cart on an air track be-
tween two springs. The sum of the forces exerted by
the two springs should at least approximately obey
Hooke’s law,
F = −kx ,
where the equilibrium point is at x = 0. The nega-
tive sign means that if the object is displaced in the
positive direction, the force tends to bring it back
in the negative direction, towards equilibrium, and
vice versa. Of course, there are no actual springs
involved in the sun or between a rock’s atoms, but
we can still learn about this type of situation in a
lab experiment with a mass attached to a spring. In
this lab, you will study how the changing velocity of
the object, in this case a cart on an air track, can
be understood using conservation of energy. Recall
that for a constant force, the potential energy is sim-
ply −F x, but for a force that is different at different
locations, the potential energy is minus the area un-
der the curve on a graph of F vs. x. In the present
case, the area formed is a triangle with base = x,
height = kx, and
1
area = base · height
2
1
= − kx2
2
(counted as negative area because it lies below the
x axis), so the potential energy is
1 2
PE = kx .
2
Conservation of energy, P E + KE = constant, gives
1 2 1
kx + mv 2 = constant .
2 2

30 Lab 11 Conservation of Energy


Preliminary Observations
You should do both of the following methods of de-
termining the spring constant.
Determining the spring constant: method 1
Pull the cart to the side with a spring scale, and
make a graph of F versus x, like the one above.
To avoid pulling at the wrong angle, it helps if you
connect the spring scale to the cart with a piece of
string. Find the combined spring constant of the two
springs, k, from the slope of the graph. (Don’t dis-
connect either spring.) Since you’ll only use method
1 as a rough check against the more precise method
2, don’t spend time taking a lot of data points.
Determining the spring constant: method 2
The second technique for determining k is to pull the
cart to one side, release it, and measure the period
of its side-to-side motion, i.e., the time required for
each complete repetition of its vibration. As we’ll
discuss later in the course, the period is nearly in-
dependent of the amount of travel, and the spring
constant is related to the period and the mass of the
cart by the equation k = m(2π/T )2 . A small pe-
riod indicates a large spring constant, since a pow-
erful spring would be required to whip the cart back
and forth rapidly. The period, T , can be found very
accurately by using a stopwatch to time many os-
cillations in a row without stopping. This method
therefore gives a very accurate value for k, which
you should use in your analysis of the conservation
of energy. Your k value from method 1 is still useful
as a check, however.
Observations
The technique is essentially the same as in lab 9,
which you may want to review. Instructions for use
of the Vernier Timer software are given in Appendix
6; you want the mode for measuring how long the
photogate was blocked. The two springs are at-
tached to the cart by sticking them directly through
the holes in the cart (not through the bumper, which
would cause the springs to drag on the track). At
the ends of the track, the springs can be attached us-
ing alligator clips, again taking care to attach them
high enough so they don’t drag.
Throughout the lab, you should only leave the air
blower turned on when you are actually using the
air track. In the past, we have burned out motors
or even melted hoses by leaving the air blowers on
continuously.
Before you start taking actual data, check whether
you have excessive friction by letting the computer
record data while the cart vibrates back and forth
a few times through the photogate. If the air track
is working right, all the time measurements should
be nearly the same, but if the data show the cart
slowing down a lot from one vibration to the next,
then you have a problem with friction. A drop in
velocity of about 1% over a half-cycle is reasonable.
The most common causes of excessive friction are
springs that are dragging on the track or springs
that are not horizontal, and thus tipping the cart
and causing one of its edges to drag.
Measure the velocity of the cart for many different
values of x by moving the photogate to various po-
sitions. Make sure you always release the cart from
rest at the same point, and when you are initially
choosing this release point, make sure that it is not
so far from the center that the springs are completely
bunched up or dragging on the track. Don’t forget
that the x you use in the potential energy should
be the distance from the equilibrium position to the
position where the vane is centered on the photogate
— if you don’t think about it carefully, it’s easy to
make a mistake in x equal to half the width of the
vane.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 What measurements besides those mentioned
above will you need to do in lab in order to check
conservation of energy?
31
P2 Find the value of x from the figure below. (I’ve
made the centimeter scale unrealistic for readability
— the real track is more than a meter long, not 14
centimeters.)
P3 Check that the units of the two methods for
finding k agree.

Self-Check
Calculate the energies at the extremes, where P E =
0 and KE = 0, and see whether the energy is staying
roughly constant. You should do this self-check as
early as possible in the lab, so that you can make
sure you’re not spending lots of time collecting data
that turn out to be bogus.

Analysis
Graph P E, KE, and the total energy as functions
of x, with error bars (see appendices 1, 2, and 3),
all overlaid on the same plot. Make sure to include
the points with KE = 0 and P E = 0. As a shortcut
in your error analysis, it’s okay if you do the error
analysis for your most typical data-point, in which
the energy is split roughly 50-50 between PE and
KE, and then assume that the same error bars on
P E, KE, and total energy apply to all the other
points on the graph as well.
Discuss whether you think conservation of energy
has been verified.

32 Lab 11 Conservation of Energy

33
12 Conservation of Momentum
Apparatus have any numerical values for the velocities, but sup-
pose we say that the initial velocity of the incoming
computer with Logger Pro software cart is one unit. Does it look like conservation of
track momentum was satisfied?
2 dynamics carts and 2 carts with magnets
1-kg weight B Mirror symmetry
1 kg slotted weight Now reenact the collision from part A, but do every-
masking tape thing as a mirror image. The roles of the target cart
2 force sensors with rubber corks and incoming cart are reversed, and the direction of
motion is also reversed.
M × +M ×
Qualitative Observations =? M × +M × ,

First you’re going to observe some collisions between What happens now? Note that mathematically, we
two carts and see how conservation of momentum use positive and negative signs to indicate the direc-
plays out. If you really wanted to take numerical tion of a velocity in one dimension.
data, it would be a hassle, because momentum de- C An explosion
pends on mass and velocity, and there would be four
different velocity numbers you’d have to measure: Now start with the carts held together, with their
cart 1 before the collision, cart 1 after the collision, magnets repelling. As soon as you release them,
cart 2 before, and cart 2 after. To avoid all this com- they’ll break contact and fly apart due to the re-
plication, the first part of the lab will use only visual pulsion of the magnets.
observations. M × +M ×
Try gently pressing the two carts together on the =? M × +M × ,
track. As they come close to each other, you’ll feel Does momentum appear to have been conserved?
them repelling each other! That’s because they have
magnets built into the ends. The magnets act like D Head-on collision
perfect springs. For instance, if you hold one cart Now try a collision in which the two carts head to-
firmly in place and let the other one roll at it, the wards each other at equal speeds (meaning that one
incoming cart will bounce back at almost exactly the cart’s initial velocity is positive, while the other’s is
same speed. It’s like a perfect superball. negative).
A Equal masses, target at rest, elastic collision M × +M ×
Roll one cart toward the other. The target cart is =? M × +M × ,
initially at rest. Conservation of momentum reads
E Sticking
like this,
Arrange a collision in which the carts will stick to-
M × +M × gether rather than rebounding. You can do this by
=? M × +M × , letting the velcro ends hit each other instead of the
where the two blanks on the left stand for the two magnet ends. Make a collision in which the target is
carts’ velocities before the collision, and the two initially stationary.
blanks on the right are for their velocities after the M × +M ×
collision. All conservation laws work like this: the =? M × +M × ,
total amount of something remains the same. You
don’t have any real numbers, but just from eye- The collision is no longer perfectly springy. Did it
balling the collision, what seems to have happened? seem to matter, or was conservation of momentum
Let’s just arbitrarily say that the mass of a cart is still valid?
one unit, so that wherever it says “M x” in the equa-
tion, you’re just multiplying by one. You also don’t

34 Lab 12 Conservation of Momentum

F Hitting the end of the track
One end of the track has magnets in it. Take one
cart off the track entirely, and let the other cart roll
all the way to the end of the track, where it will
experience a repulsion from the fixed magnets built
into the track. Was momentum conserved? Discuss
this with your instructor.
G Unequal masses
Now put a one kilogram mass on one of the carts,
but leave the other cart the way it was. Attach the
mass to it securely using masking tape. A bare cart
has a mass of half a kilogram, so you’ve now tripled
the mass of one cart. In terms of our silly (but con-
venient) mass units, we now have masses of one unit
and three units for the two carts. Make the triple-
mass cart hit the initially stationary one-mass-unit
cart.
3M × +M ×
=? 3M × +M × , each other, and record the forces on the sensors. The
These velocities are harder to estimate by eye, but if
you estimate numbers roughly, does it seem possible
that momentum was conserved?
Quantitative Observations
Now we’re going to explore the reasons why momen-
tum always seems to be conserved. Parts H and I
will be demonstrated by the instructor for the whole
class at once.
Attach the force sensors to the carts, and put on the
rubber stoppers. Make sure that the rubber stoppers
are positioned sufficiently far out from the body of
the cart so that they will not rub against the edge
of the cart. Put the switch on the sensor in the
+10 N position. Plug the sensors into the DIN1
and DIN2 ports on the interface box. Start up the
Logger Pro software, and do File>Open>Probes and
Sensors>Force Sensors>Dual Range Forrce>2-10 N
Dual Range. Tell the computer to zero the sensors.
Try collecting data and pushing and pulling on the
rubber stopper. You should get a graph showing how
the force went up and down over time. The sensor
uses negative numbers (bottom half of the graph) for
forces that squish the sensor, and positive numbers
(top half) for forces that stretch it. Try both sensors,
and make sure you understand what the red and blue
traces on the graph are showing you.
H. Put the extra 1-kilogram weight on one of the
carts. Put it on the track by itself, without the other
cart. Try accelerating it from rest with a gentle,
steady force from your finger. You’ll want to set the
collection time to a longer period than the default.
Position the track so that you can walk all the way
along its length (not diagonally across the bench).
Even after you hit the Collect button in Logger Pro,
the software won’t actually start collecting data until
it’s triggered by a sufficiently strong force; squeeze
on one of the sensors to trigger the computer, and
then go ahead and do the real experiment with the
steady, gently force.
What does the graph on the computer look like?
I. Now repeat H, but use a more rapid acceleration
to bring the cart up to the same momentum. Sketch
a comparison of the graphs from parts H and I.
Discuss with your instructor how this relates to mo-
mentum.
J. You are now going to reenact collision A, but don’t
do it yet! You’ll let the carts’ rubber corks bump into
carts will have equal mass, and both forces will be
recorded simultaneously. Before you do it, predict
what you think the graphs will look like, and show
your sketch to your instructor.
Switch both sensors to the +50 N position, and open
the corresponding file on the computer.
Zero the sensors, then check the calibration by bal-
ancing a 1 kg weight on top, taking data, zooming
in, and putting the mouse cursor on the graph. You
will probably find that the absolute calibration of
the sensor is very poor when it’s used on the 50 N
scale; keep this in mind when interpreting your re-
sults from the collision.
Now try it. To zoom in on the relevant part of the
graph, use the mouse to draw a box, and then click
on the magnifying glass icon. You will notice by
eye that the motion after the collision is a tiny bit
different than it was with the magnets, but it’s still
pretty similar. Looking at the graphs, how do you
explain the fact that one cart lost exactly as much
momentum as the other one gained? Discuss this
with your instructor before going on.
K. Now imagine – but don’t do it yet – that you
are going to reenact part G, where you used unequal
masses. Sketch your prediction for the two graphs,
and show your sketch to your instructor before you
go on.
Now try it, and discuss the results with your instruc-
tor.
35
13 Conservation of Momentum in Two Di-
mensions
Apparatus
photogate (PASCO) . . . . . . . . . . . . . . . . . . . . . . 1/group
computer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
small steel and plastic balls of various masses 2/group
plastic rulers
protractor
scotch tape

Goal
Test whether momentum is conserved in a collision
of two balls.

Introduction
Pool players have an intuitive feeling for conserva-
tion of momentum: they can visualize the results of
a collision of two pool balls in advance. They also
know that certain shots are impossible. For instance,
there is no way to make the cue ball bounce back di- balls using the photogate and the computer. The
rectly from a collision with another ball (except by use of the photogate and the computer software that
putting spin on it, which creates an external friction works with it is explained in Appendix 6. Since on
force with the felt). They understand that the angles any given trial you can only use the photogate to
are important, so without knowing it, they are doing measure the speed of a single ball, you will have to
mental estimates involving momentum as a vector: reproduce the collision at least three times to mea-
a quantity that has both magnitude and direction. sure the three speeds involved. Actually, you will
want to measure each of the three speeds several
In this lab, you will be studying collisions similar to times in order to get a good estimate of your ran-
the collision of the cue ball with an initially station- dom errors.
ary ball. One of the basic principles involved is the
conservation of momentum:
The Principle of Conservation of Momentum
No matter how a set of objects interact with each
other, as long as no external force is present, the
vector sum of their momenta is conserved. That is,
p1i + p2i + . . . = p1f + p2f + . . ., where the “. . . ”
means that if there are more than two objects in-
volved, all their momenta should be added like vec-
tors.

The technique
The idea is to set up an off-center collision, as shown
below, and measure the initial and final speeds of the
To reproduce the same initial speed for the projectile
(ball 1), you can build a little ramp out of two plastic
rulers taped together at a 90-degree angle. A block
of wood can be taped in the ramp at the top to keep

36 Lab 13 Conservation of Momentum in Two Dimensions

them braced. The block of wood also serves as a from the collision should you place it? It should
convenient reference point: you can release the ball be as close as possible to the collision, because the
from the point where it touches the block. balls do gradually slow down as they roll, and you
want to know the speeds immediately before and
You should choose a completely asymmetrical setup:
after the collision. However, the balls bounce a little
two balls of different masses, and a collision in which
immediately after the collision, so don’t put the it
the projectile does not hit the target head-on.
so close to the collision that they are still bouncing
It is critical that you position the target ball at ex- when they go through it.
actly the same place every time. Marking the table
and placing the ball on the mark is not good enough.
The best technique is to put a piece of scotch tape Prelab
on the table and use a ball-point pen to make a tiny
The point of the prelab questions is to make sure
impression in it for the target ball to sit in.
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
Tips you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
You want to avoid conditions for which any of the you’re just setting yourself up for failure in lab.
speeds involved are too slow, because then the balls P1 Draw an example of a collision, showing the
tend to be accelerated, decelerated, or deflected by balls before and after it happens, in which |p1i | =
tiny bumps in the tabletop. If you notice the balls 0.020 kg · m/s, |p1f | = 0.010 kg · m/s, and |p2f | =
wandering and wavering as they roll, they are go- 0.010 kg · m/s, but momentum was not conserved.
ing too slow. Generally speaking, sufficiently high (As in the actual lab, the target ball starts at rest.)
speeds are achieved if the ramp is at least 7 cm high. Explain.
Using the heavier ball as the projectile helps to keep
the final speeds high. P2 If the magnitude of the initial momentum is the
same as the magnitude of the total final momentum,
A good way to test whether your speeds are sufficient does that mean momentum was conserved?
is to measure the angles at which the balls emerge
from the collision, and see if they are the same every
time, to within a tolerance of 5-10 degrees. If the Self-Check
angles are not reproducible to this level of variation,
then the balls are not going fast enough. Analyze your data without error analysis, and make
sure your graphical and analytical results are the
You will want to use vernier calipers to measure the same. Check whether momentum appeared to be at
diameters of the balls. Ask your instructor for help least approximately conserved.
if you don’t know how to read a vernier scale.
Note that at the instant of collision, the balls are
touching, but their centers are not at the same point. Analysis
This means you have to be careful about how you Test whether momentum was conserved, doing your
measure the angles. vector addition once using the analytic method and
If you did not position the photogate at the height of once using the graphical method. Take into account
the center of the ball whose speed you wanted to de- the random errors in your measurements.
termine, then the computation of the ball’s speed be-
comes complicated — don’t just divide the diameter
of the ball by the time from the computer. Discuss
this with your instructor once you have a working
setup.
You should have opposite signs for the components
of the balls’ final momenta in the direction perpen-
dicular to the projectile’s original direction of mo-
tion.
You will be putting the photogate in three different
positions to measure the three velocities. How far

37
14 Torque
Apparatus Construct a setup like the one shown above. Avoid
any symmetry in your arrangement. There are four
meter stick with holes drilled in it . . . . . . . . 1/group forces acting on the meter stick:
spring scales, calibrated in newtons
FH = the weight hanging underneath
weights
string FM = Earth’s gravity on the meter stick itself
protractors FL = tension in the string on the left
hooks FR = tension in the string on the right

Each of these forces also produces a torque.

Goal
Test whether the total force and torque on an object
at rest both equal zero.
Introduction
It is not enough for a boat not to sink. It also must
not capsize. This is an example of a general fact
about physics, which is also well known to people
who overindulge in alcohol: if an object is to be in
a stable equilibrium at rest, it must not only have
zero net force on it, to keep from picking up momen-
tum, but also zero net torque, to keep from acquiring
angular momentum.
In order to determine whether the total force is zero,
you will need enough raw data so that for each torque
you can extract (1) the magnitude of the force vec-
tor, and (2) the direction of the force vector. In order
to add up all the torques, you will have to choose an
axis of rotation, and collect enough raw data to be
able to determine for each force (3) the distance from
the axis to the point at which the force is applied to
the ruler, and (4) the angle between the force vector
and the line connecting the axis with the point where
the force is applied. Note that the meter stick’s own
weight can be though of as being applied at its center
of mass.
You have a selection of spring scales, so use the right
one for the job — don’t use a 20 N scale to measure
0.8 newtons, because it will not be possible to read
it accurately. If you need to swap in a new spring
scale, don’t forget to calibrate it.
Since the analysis requires you to compute the to-
tal torque a second time using a different choice of
axis, you cannot neglect to measure any of the angles
involved.

Observations
Weigh your meter stick before you do anything else;
they don’t all weigh the same amount.
For each spring scale, hang a known weight from it,
and adjust the calibration tab so that the scale gives
the correct result.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 You have complete freedom in defining what
point is to be considered the axis of rotation — if
one choice of axis causes the total torque to be zero,
then any other choice of axis will also cause the to-
tal torque to be zero. It is possible to simplify the
analysis by choosing the axis so that one of the four
torques is zero. Plan how you will do this.

38 Lab 14 Torque
P2 All the torques will be tending to cause rota-
tion in the same plane. You can therefore use plus
and minus signs to represent clockwise and counter-
clockwise torques. Choose which one you’ll call pos-
itive. Using your choice of axis, which of the four
torques, τH , τM , τL , and τR , will be negative, which
will be positive, and which will be zero?
P3 Suppose that in the figure above, the angle be-
tween the meter stick and the hanging weight is 80 ◦ ,
the mass of the hanging weight is 1 kg, and the mass
of the meter stick is 0.1 kg. If a student is then try-
ing to calculate the x components of the forces FM
and FH , why is it incorrect to say

FM ,x = (0.1 kg)(9.8 m/s2 )

and

FH,x = (1 kg)(9.8 m/s2 )(cos 80 ◦ )?

Analysis
Determine the total force and total torque on the
meter stick. For the forces, I think a graphical cal-
culation will be easier than a numerical one.
Finally, repeat your calculation of the total torque
using a different point as your axis. Although you’re
normally expected to do your analysis completely in-
dependently, for this lab it’s okay if you find the total
torque for one choice of axis, and your lab partners
do the calculation their own choices.
Error analysis is not required. For extra credit, you
can do error analysis for one of your total torques.

39
15 The Moment of Inertia
Apparatus
meter stick with hole in center . . . . . . . . . . . . 1/group
nail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1/group
fulcrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1/group
slottedmass set . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
duct tape
sliding bracket to go on
meter stick . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
U-shaped hook for hanging
weights from bracket . . . . . . . . . . . . . . . . . . . . . 1/group
computer Vernier software
1/group photogate and adapter box . . . . . . .1/group
triple-beam balance . . . . . . . . . . . . . . . . . . . . . . .1/group
Goal
Test the equation τtotal = Iα, which relates an ob-
ject’s angular acceleration to its own moment of in-
ertia and to the total torque applied to it.
Introduction
Newton’s first law, which states that motion in a
straight line goes on forever in the absence of a force,
was especially difficult for scientists to work out be-
cause long-lasting circular motion seemed much more
prevalent in the universe than long-lasting linear mo-
tion. The sun, moon and stars appeared to move
in never-ending circular paths around the sky. A
spinning top could continue its motion for a much
longer time than a book sliding across a table. Ren
Descartes (b. 1596) came close to stating a law of in-
ertia like Newton’s, but he thought that matter was
made out of tiny spinning vortices, like whirlwinds
of dust. Galileo, who among Newton’s predecessors
came closest to stating a law of inertia, was also con-
fused by the issue of circular versus linear inertia. An
advocate of the Copernican system, in which the ap-
parent rotation of the sun, stars and moon was due
to the Earth’s rotation, he knew that the apparently
motionless ground, trees, and mountains around him
must be moving in circles as the Earth turned. Was
this because inertia naturally caused things to move
in circles?
Newton, like other giants of science, saw how to focus
on the simple rather than the complex. His law of
40 Lab 15 The Moment of Inertia
inertia was completely linear. In his view, all the
common examples of circular motion really involved
a force, which kept things from going straight. In
the case of a spinning top, for instance, Newton (a
confirmed atomist) would have visualized an atoms
in the top as being acted on by some kind of sticky
force from the other atoms, which would keep it from
flying off straight. Linear motion was the simplest
type, needing no forces to keep it going. Circular
motion was more complex, requiring a force to bend
the atoms’ trajectories into circles.
Even though circular motion is inherently more com-
plicated than linear motion, some very close analo-
gies can be made between the two in the case where
an object is spinning rigidly. (An examples of non-
rigid rotation would be a hurricane, in which the
inner parts complete a rotation more rapidly than
the outer parts.) In analogy to Newton’s first law,
Ftotal = ma, we have
τtotal = Iα ,
where the angular acceleration α replaces the linear
acceleration a, the total torque plays the role given
to the total force, and the moment of inertia I is
used instead of the mass. In this lab, you are going
to release an unbalanced rotating system — a meter
stick on an axle with weights attached to it — and
measure its angular acceleration in response to the
nonzero gravitational torque on it.
Every equation you learned for constant acceleration
can also be adapted to the case of rotation with con-
stant angular acceleration, simply by translating all
the variables. For instance, the equation vf2 = 2ax
for an object accelerating from rest can be translated
into the valid rotational formula ωf2 = 2aθ.
mr2 ,
P
The moment of inertia is defined as I =
where m can be thought of as the mass of an indi-
vidual atom comprising the rotating body, and r is
the distance of that atom from the axis of rotation.
The word “moment” in “moment of inertia” does
not refer to a moment in time, but is used instead
in a more old-fashioned sense of “importance” or
“weight,” as in “matters of great moment.” The
idea is that the factor of r2 gives more importance
to the an atom that is far from the axis of rotation.
Because the symbol I is used, there is a tendency
for students to refer to it as “inertia,” but inertia
is a different and nonquantitative concept, referring
to the tendency of objects to stay at rest or stay in
motion.
In practice, it is not practical to carry out a sum over
all the atoms. The object whose rotation you will
study in this lab will consist of a meter stick pivot-
ing at its center, with various weights hanging from
it in various places. Both the hanging weights and
the meter stick itself will contribute to the moment
of inertia. To a good approximation, each hanging
weight can be treated as if all its atoms were con-
centrated at its center. Calculus can also be used
to derive formulae for the moments of inertia of ob-
jects of various shapes, such as a sphere, a cylinder
rotating along its axis, etc. One such formula is
I = (1/12)M L2 for the moment of inertia of a rigid
rod rotating about an axis passing perpendicularly
through its center. You can use this formula as a
good approximation for the meter stick’s contribu-
tion to the moment of inertia, with L = 1 m.
Preliminaries
The meter stick is supported on the fulcrum via a
nail through the hole in its center. You want to start
by producing a balanced arrangement of weights at-
tached to the meter stick, as in figure (a) below. The
idea is that if you first balance this configuration
carefully, then you know that the net gravitational
torque on it is zero. If you then hang another weight
from the previously empty hanger, as in (b), then
you know that the total torque simply equals the
torque produced by the earth’s gravitational force
on the added weight.
For ease of adjustment, you can use duct tape, wrapped
sticky-side-out, to attach the slotted weights to the
meter stick. You can then balance your initial config-
uration simply by sliding the weights around. Do not
choose a symmetric setup, i.e., use unequal weights.
The masses need to be slid to the left and right in or-
der to achieve equilibrium, but it is less obvious that
it also makes a difference how high the weights are
placed. That is, the center of mass of the whole bal-
anced setup must coincide both vertically and hori-
zontally with the nail. The concept is shown in the
figure above using a rectangle in place of the actual
apparatus. In (d), there will always be a clockwise
torque on the rectangle, because the center of mass
is to the right of the nail.
In (e), there is zero torque if the rectangle is ini-
tially released from this horizontal position, but the
equilibrium is unstable, because its center of mass is
above the axis of rotation. Our experiment depends
on the cancellation of the gravitational torques on
everything but the extra weight, but in a case like
(e), this assumption would only be valid when the
apparatus was initially released from horizontal. Later
in the motion, there would be an undesired and un-
known extra torque. Although it is visually obvious
in this figure that the rectangle’s center of mass is
too high, you can’t tell visually with the actual ap-
paratus. The way to tell if the center of mass is too
high is that if you tilt the meter stick a little bit
to the right, it immediately accelerates clockwise,
whereas if you tilt it a little to the left, it accelerates
counterclockwise.
In (f), we have a stable equilibrium. Again, there is
an unknown, undesired torque unless the rectangle
just happens to be horizontal. You can tell if you
have this situation because the apparatus can swing
back and forth about its stable equilibrium position.
You want a neutral equilibrium, i.e., no matter what
angle you release it from, the meter stick just stays
there.
41
Observations
Now add the extra weight so that the meter stick
is slightly unbalanced. The idea of this lab is to
release the meter stick and use the photogate to find
how quickly it is moving once it has rotated through
some angle, using the photogate to find the amount
of time required for the tip of the meter stick to pass
through the photogate. From your measurement of
∆t using the photogate, you can find ω = ∆θ/∆t,
which is an approximation to the meter stick’s final
angular velocity. Instructions for using the computer
software are given in appendix 6; you want the mode
for measuring how long the photogate was blocked.
Once you know the meter stick’s final value of ω,
you can extract the angular acceleration. This can
then be compared with the theoretical value of the
angular acceleration from τtotal = Iα.
Tips:
You may want to put something under the ful-
crum base to raise everything up higher.
Although the balanced configuration, with τtotal
0, still has τtotal = 0 no matter what angle it
is at, the torque exerted by the extra weight
does depend a little on what angle the meter
stick is at. This is because of the factor of
sin θ in the definition τ = rF sin θ. Since the
torque is not constant, the angular accelera-
tion is not constant, leading to complications.
You can avoid this problem by confining all
your measurements to a fairly small range of
positions near horizontal. As long as θ is fairly
close to 90 ◦ , sin θ is extremely close to 1, and
it is a good enough approximation to assume a
constant torque rF producing a constant an-
gular acceleration. For instance, as long as θ
is within 20 ◦ above or below horizontal, sin θ
changes by no more than 0.06.
Although you want to work only with nearly
horizontal positions of the meter stick so that
the torque is approximately constant, you also
need to make sure that the total angle tra-
versed by the meter stick is still reasonably
large compared to the angle traversed while the
meter stick is blocking the photogate. Other-
wise your measurement of ω = ∆θ/∆t will not
be a good approximation to the final instanta-
neous angular velocity.
As you will find in your prelab, the angular ac-
celeration depends on the square of the angle
42 Lab 15 The Moment of Inertia
∆θ. Measuring this angle accurately is there-
fore vital in order to get a good result. A
protractor cannot measure an angle this small
with sufficient accuracy. Use trigonometry to
determine this angle.
It’s easiest if you use radian measure through-
out. The equation τtotal = Iα is only true if a
is measured in radians/s2 .
The sliding bracket and hook contribute both
to the total torque and the moment of inertia,
so you’ll have to weigh them.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Derive an equation for the experimental value
of the angular acceleration, expressed in terms of
= quantities you will actually measure directly, includ-
ing the quantities θ and ∆θ defined in the figure be-
low. Note that this lab is exactly analogous to the
previous lab where you found a linear acceleration
using a similar setup.
P2 Why would it not be meaningful to try to deal
with the meter stick’s velocity, rather than its angu-
lar velocity?
Self-Check
Do all your analysis in lab.
Analysis
Extract theoretical and experimental values of the
angular acceleration from your data, and compare
them.
No analysis of random errors is required, because the
main source of error is the systematic errors arising
from friction and the various approximations, such
as the assumption that sin θ is approximately equal
to 1.

43
16 Absolute Zero
Note to the lab technician: The dessicant needs to
be dry before the experiment. If it’s blue, it’s dry. If
it’s pink, it needs to be pumped on for a few hours
with a vacuum pump while heating it with a hair
dryer. Also, please put 2 liters of alcohol in a freezer
overnight, then put them in the ice chest.
Apparatus
gas capillary tube
large test tube
mercury thermometer
glass syringe
electric heating pad
oven mitts
latex tubing
ice
string
funnels
clamps
gripper clamps
Introduction
If heat is a form of random molecular motion, then
it makes sense that there is some minimum temper-
ature at which the molecules aren’t moving at all.
With fancy equipment, physicists have gotten sam-
ples of matter to within a fraction of a degree above
absolute zero, but they have never actually reached
absolute zero (and the laws of thermodynamics ac-
tually imply that they never can). Nevertheless, we
can determine how cold absolute zero is without even
getting very close to it. Kinetic theory tells us that
the volume of an ideal gas is proportional to how
high it is above absolute zero. In this lab, you’ll
measure the volume of a sample of air at tempera-
tures between 0 and 100 degrees C, and determine
where absolute zero lies by extrapolating to the tem-
perature at which it would have had zero volume.
Observations
Tie a short piece of string to the thermometer so
that you’ll be able to pull it back out of the beaker
when you want to without dipping your hands in
hot water. Start heating the water up to the boiling
44 Lab 16 Absolute Zero
point. (If you leave the thermometer in the water
while it’s heating, you’ll be able to observe later the
interesting fact that the water stops heating up once
it reaches the boiling point.) If the water starts boil-
ing before you’re ready, just turn off the heat and
reheat it later – it doesn’t cool off very fast.
The capillary tube is sealed at the bottom and open
at the top, with a large bulb full of dessicant just
below the top to keep the air inside dry. There is
a small amount of mercury inside the tube. Right
now, the mercury is probably “floured,” i.e., broken
up into small pieces sticking on the sides of the tube.
The idea is to collect the mercury into a single drop,
with a sample of air trapped in the capillary tube
under it. The mercury simply acts as a seal. As the
air is heated and cooled, it expands and contracts,
and you can measure its volume by watching the
mercury seal rise and drop. By the way, don’t be
scared of the mercury; mercury vapor is a deadly
poison, but liquid mercury is entirely harmless unless
you ingest it or get it in an open cut. There is a small
filter made of glass wool at the top end of the bulb,
which will keep the mercury from getting out.
Remove the gas syringe from the box, being care-
ful not to let the glass plunger drop out and break.
Connect it to the capillary tube with a piece of tub-
ing.
First you need to get the mercury into a single blob
in the cavity at the top of the capillary, where it
widens out just below the bulb. If it’s already form-
ing a seal across the capillary tube, you won’t be
able to get it to move, because it’s trapped between
the pressures of the inside air and the outside air.
You can break the seal by opening the stopcock and
drawing some air out with the syringe. (Note that
the stopcock has three holes; two are lined up with
the knob, and the third one is on the side marked
with a dot on the knob.) If this doesn’t break the
seal, you can very gently shake the capillary tube
longitudinally. Shaking it longitudinally means mov-
ing it along its own length, like an arrow. Don’t tap
it, and don’t shake it transversely (like a fever ther-
mometer or a magician’s wand), because students
have broken tubes that way in the past. Now dis-
connect the tube from the syringe, and, if necessary,
shake it extremely gently upside-down to get all the
mercury droplets to collect in the cavity.
At this point, if you put the tube upright again, the
mercury drop will sit at the very top of the capil-
lary, with a sample of air trapped below it filling the
entire tube. This is no good, because most of the
temperatures you’ll be using in this lab are hotter
than room temperature, so you need room for the
air sample to expand without forcing the mercury
out into the cavity. Here’s how to get a smaller vol-
ume of air trapped under the mercury. Push the
plunger all the way into the syringe, open the stop-
cock, and connect the syringe to the tube, leaving
the tube horizontal with the mercury in the cavity.
Now pull the plunger out until you’ve created a 40%
vacuum. If you have the stopcock in the correct po-
sition, it should take quite a strong force to pull the
plunger out this far. Now bring the tube upright
again, and gently allow the plunger to slide back in.
At this point, the mercury should be about 40% be-
low the top of the capillary, and you can disconnect
the syringe.
Detach the syringe and tubing, so from now on, ev-
erything is always at constant pressure! We want
temperature and volume to be the only variables
that change in this experiment. By leaving every-
thing open to the air in the room, we guarantee that
the pressure will equal the air pressure in the room.
If necessary, bring the water back to a boil, and then
turn off the heat again. Being careful not to burn
yourself, lower the clamp so the test tube is almost
touching the tabletop; this way, if it slips out of the
clamp, it won’t fall far enough to break. (I broke one
of the test tubes myself by letting it slip this way.)
Insert the thermometer and the capillary tube, and
give them a minute or so to come to equilibrium with
the water.
You can now start taking a series of temperature and
volume measurements as the water in the test tube
gradually cools down towards room temperature.
The cooling process is rapid at first. If you get im-
patient, you can gently pour a small amount of cool
water in the top, making sure to let it equilibrate for
a few minutes afterward before taking data. Don’t
try to swirl the test tube around in order to speed
up the equilibration – that’s what I was trying to do
the time the test tube slipped out of the clamp and
broke.
When the water gets close to room temperature, the
cooling process slows down. At some point, you may
wish to fill a beaker with lukewarm water and im-
merse the end of the flask in it in order to speed up
the cooling. Meanwhile, you can start preparing a
sample of very cold alcohol mixed with ice.
Once you have data at temperatures down to near
room temperature, pour some liquid off of the alcohol-
ice mixture, and use it to replace the water in the
flask. Make sure you don’t get ice in the flask, which
makes it impossible to insert the capillary tube and
thermometer.
Analysis
Graph the temperature and volume against each other.
Does the graph appear to be linear? If so, extrap-
olate to find the temperature at which the volume
would be zero.
If your data are nice and linear, then your main
source of error will be random errors, and you should
then determine error bars for your value of absolute
zero using the techniques discussed in Appendix 4.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Should you measure the volume from the top,
the middle, or the bottom of the mercury? Explain.
45
17 The Clement-Desormes Experiment
Apparatus
large flask
glass syringe
water manometer
helium (medium size cylinder, $40 from Party City)
difluoroethane (sold in cans as gas duster at Fry’s)
stopwatch
hose clamps
grabber clamps
stands

Introduction
Although the theory that matter was made of atoms
started to be talked about seriously by scientists as
duction is a slow process, we can treat this as insu-
early as Galileo’s time, scientists generally didn’t
lated expansion, as discussed in Appendix 2 of Sim-
think of it as something that was literally true. They
ple Nature. If the gas is a monoatomic one, such
considered the atomic theory to be a useful model,
as helium, then the amount of cooling of the gas, as
but they thought that any fundamental explanation
proved in the book, is given by the relation T ∝ P b ,
of real-world phenomena should avoid talking about
where b = 2/5. If the gas is not monoatomic, how-
hypothetical things like atoms. This feeling was so
ever, then its molecules can rotate,1 and at any given
strong that the physicist Ludwig Boltzmann, who
time some of its energy is in the form of kinetic en-
came up with an atomic explanation of entropy, was
ergy along the x, y, and z axes, but some is in the
driven to suicide by the harsh criticism to which his
form of rotational kinetic energy. Extracting a given
ideas were subjected. Even more suspect than the
amount of energy from a diatomic or polyatomic gas,
existence of atoms was any attempt to discuss things
therefore, doesn’t cool it as much as it would cool a
like the shapes of molecules that could be formed
monoatomic gas, and it turns out that b = 2/7 for a
by putting them together like tinkertoys; such ideas
diatomic gas, and 1/4 for a polyatomic gas.2
seemed much too far removed from the possibility of
any experimental testing. Wait one minute for the air to warm back up to room
temperature. The pressure comes back up somewhat
Surprisingly, then, a simple experiment, due to Clem-
as the air warms back up, and although you should
ent and Desormes, is capable of distinguishing two
wait a full minute to make sure it’s back in thermal
samples of gas that differ only by the shape of their
equilibrium, most of the rewarming occurs during
molecules, even if the gases have the same density
the first few seconds after you finish venting the ini-
and are composed of molecules having the same mass.
tial pressure. The pressure will recover to a value
Use the glass syringe to apply a slight overpressure p2 which is less than p1 . The ratio p2 /p1 gives the
to the air inside the flask, causing the difference in value of b for the gas.3
height between the water in the two sides of the
I’m still working on improving this lab. The ba-
manometer to be about 30 cm. Wait one minute
sic idea I have in mind is to have you do the lab
to make sure the air is in thermal equilibrium with
once with helium (monoatomic), air (diatomic), and
the room, and then take a pressure reading, p1 . Re-
lease the pressure by popping the cork for precisely 1 An individual atom in a monoatomic gas has essentially

one second, timed on a stopwatch. The air cools
slightly due to its expansion, because it does me-
chanical work as it exits throught the valve. How-
ever, because the expansion is rapid, and heat con-
all its mass concentrated in the nucleus exactly at its center,
so it takes an effectively infinite amount of energy to make it
rotate with a certain amount of angular momentum.
2 You’ll often see this stated in terms of the variable γ =
1/(1 − b), which takes on the values 5/3, 7/5, and 4/3.
3 In terms of the variable γ, we have γ = p /(p − p ).
1 1 2

46 Lab 17 The Clement-Desormes Experiment

difluoroethane (polyatomic), and observe the differ-
ences in the results due to the different shapes of
the molecules. There are various systematic errors
in the experiment, so my own absolute results for
the b of air haven’t been of extremely high preci-
sion; however, in a comparative experiment, I think
it will be easy to see a difference in b between the
gases. One possible problem with the air is that it
contains water vapor, which messes up the thermo-
dynamic properties of the air, because water droplets
can condense out of the air when the pressure is
dropped suddenly, as when you open a can of beer.
The helium and difluoroethane shouldn’t have this
problem. In the spring semester of 2008, we tried
all three gases, and found that it was fairly easy to
detect a clear systematic difference between a higher
b for air (.20, .29, .33, .29, and .31 for the five lab
groups) and a lower one for difluoroethane (.18, .20,
.33, .25, and .24), but the results for helium were
much lower than theory, and barely distinguishable
from air (.29, .35, .35, .31, and .31, versus .40 ac-
cording to theory). This may be because we’re not
actually getting the flasks as full of pure helium as
we think we are.
Some of the flasks have holes at both the top and the
bottom. With these flasks, it’s a good idea to intro-
duce the helium through the bottom hole, since it’s
lighter than air, and will rise. The difluoroethane,
on the other hand, should be put in through the top
hole, because it’s heavier than air. I don’t know if
it will be practical to use the helium with the flasks
that only have holes at the top.
Both the helium and the difluoroethane can displace
the beathable air in the classroom, and the amount
of helium in the large canister is particularly big. For
this reason, I’ve been dispensing the helium outside
the classroom.
The difluoroethane is a liquid when it’s pressurized
inside the can. When you vent some of the pressure
through the nozzle, the pressure drops, and some of
it vaporizes and comes out. The vaporization con-
sumes energy, so the can becomes cold. If you hold
the can upside down and spray it, liquid is emitted
rather than gas; this liquid is extremely cold, and
can cause frostbite if it gets on your skin. The gas is
not flammable, and does not harm the ozone layer.
Some teenagers have intentionally inhaled it to get
high, so the manufacturers have added a bitterant.

47
18 The Pendulum
Apparatus Notation and Terminology
string When a moving thing, such as a wave, an orbit-
cylindrical pendulum bobs ing planet, a wheel, or a pendulum, goes through
hooked masses a repetitive cycle of motion, the time required for
protractor one complete cycle is called the period, T . Note
stopwatch that a pendulum visits any given point once while
computer with photogate and Vernier Timer soft- traveling in one direction and once while traveling
ware in the opposite direction. The period is defined as
clamps (not hooks) for holding the string how long it takes to come back to the same point,
tape measures traveling in the same direction.
meter sticks

Goal
Find out how the period of a pendulum depends on
its length and mass, and on the amplitude of its
swing. From a to g is one full period of the pendulum. From a
to e is not a full period. Even though the pendulum has
returned at e to its original position in a, it is moving in the
Introduction opposite direction, and has not performed every type of
motion it will ever perform.
Until the industrial revolution, the interest of the
world’s cultures in the measurement of time was al-
most entirely concentrated on the construction of
calendars, so that agricultural cycles could be an-
ticipated. Although the Egyptians were the first to
divide the day and night into 12 hours, there was no
technology for measuring time units smaller than a
day with great accuracy until four thousand years
later.
Galileo was the first to realize that a pendulum could
be used to measure time accurately — previously, he
had been using his own pulse to measure the time re-
quired for objects to roll down inclined planes. The
legend is that the idea came to him while he watched
The amplitude of a repetitive motion is a way of
a chandelier swinging during a church service. Sen-
describing the amount of motion. We can define
tenced to house arrest for suspicion of heresy, he
the amplitude, A, of the pendulum’s motion as the
spent the last years of his life trying to build a more
maximum angle to which it rises, i.e., half the total
practical pendulum clock that would run for long
angle swept out. Let us denote the mass of the bob,
periods of time without tending. This technical feat
or weight at the end of the pendulum, by m, and
was only achieved later by Christian Huygens. Along
the length of the pendulum, from the pivot to the
with the Chinese invention of the compass, accurate
middle of the weight, as L.
clocks were vital for European exploration by sea,
because longitude can only be determined by astro-
nomical observations combined with accurate mea- Observations
surements of time.
Make observations to determine how the period, T ,
depends on A, L, and m. You will want to use the

48 Lab 18 The Pendulum

technique of isolation of variables. That means that
rather than trying many random combinations of
A, L, and m, you should keep two of them constant
while measuring T for various values of the third
variable. Then you should shift your attention to
the next variable, changing it while keeping the other
two constant, and so on. Be sure to try quite a few
values of the variable you are changing, so you can
see in detail how T depends on each variable.
The period can be measured using the photogate.
See appendix 6 for how to use the computer soft-
ware; you want the mode that’s meant specifically
for measuring the period of a pendulum. Note that
the bob is what is blocking the photogate, so if your
bob is irregularly shaped, your measurements could
be messed up if it changed orientation between one
pass through the photogate and the next. The eas-
iest way to make sure this problem doesn’t occur is
to use a bob with a circular cross-section, so it has
the same width no matter which way the photogate
cuts through it.
One of the notable differences between the way stu-
dents and professional scientists approach experi-
ments is that students tend to be timid about explor-
ing extreme conditions. In this experiment, there is
a big advantage to taking measurements over wide
ranges of each of the three parameters, because it
may be impossible to ascertain how the period de-
pends on a parameter if you only explore a small
range. When changing L, you can go up to four
meters if you hang the pendulum from the balcony;
however, you should avoid lengths so short that they
are comparable to the size of the bob itself, since
such short lengths would have anomalous behavior.
For large values of L, it’s not practical to use the
computer, so use a stopwatch instead. Don’t just
time one oscillation, because then the precision of
your timing will be horrible. Measure the time re-
quired for some large number of oscillations.
Warning: Since L is measured to the middle of the
weight, you must change the length of the string
if you want to vary m while keeping L constant,
compensating for the different physical size of the
new weight.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 What is the maximum possible amplitude for
a pendulum of the type you’ll use, whose bob hangs
from a string? If you were using a pendulum with a
stiff rod instead of a string, you could release it from
straight up. What would its period be if you could
release it from exactly straight up?
P2 Referring to appendix 5, how will you tell from
your log-log plot whether the data follow a power
law, i.e., whether it is even appropriate to try to
extract p? (If you’ve already done lab 6, it’s exactly
the same technique.)
Self-Check
Figure out which variable T depends on most strongly,
and extract p (see below).
Analysis
Graph your data and state your conclusions about
whether T depends on A, L and m. Remember
that on a graph of experimental data, the horizontal
axis should always be the quantity you controlled di-
rectly, and the vertical axis should be the quantity
you measured but did not directly select. The pho-
togate is so accurate that there is not much point
in putting error bars on your graph — they would
be too small to see. Remember, however, that there
are some fairly significant systematic errors, e.g., it is
hard to accurately keep L the same when switching
masses.
It may happen that when you change one of the vari-
ables, there are only small, insignificant changes in
the period, but depending on how you graph the
data, it may look like these are real changes in the
period. Most computer graphing software has a de-
fault which is to make the y axis stretch only across
the range of actual y data. e.g., if your periods
were all between 0.567 and 0.574 s, then the soft-
ware makes an extremely magnified graph, with the
y axis running only over the short range from 0.567
to 0.574 s. On such a scale, it may seem at first
glance that there are some major changes in the pe-
riod. To help yourself interpret your graphs, you
should make them all with the same y scale, going
from zero all the way up to the highest period you
ever measured. Then you’ll be comparing all three
graphs on the same footing.
Of the three variables, find the one on which the
period depends the most strongly, and use the tech-
49
niques outlined in appendix 5 to see if you can find
an equation describing the relationship between the
period and that variable. Assume that the equation
is of the form

T = cxp ,

where x would actually be A, L or m, and c and p

are constants. The constant p is important, and is
expected to be the same for all pendula. For in-
stance, if you find that the mass is the variable that
has the greatest effect on the period, and that the
relationship is of the form T = cm3 , then you have
discovered something that is probably generally true
for all pendula: that the period is proportional to the
cube of the mass. The constant c is just some bor-
ing number that’s not worth extracting from your
graphs; it’s the exponent p that’s interesting and
universally valid.

50 Lab 18 The Pendulum

51
19 Resonance
Apparatus
vibrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
stopwatch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
multimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
banana plug cables
Thornton power supplies (in lab benches in 416)
Leybold 521 545 17-volt DC power supplies1
24 V AC power supplies

Goals
Observe the phenomenon of resonance.
Electrical setup, top view.
Investigate how the width of a resonance de-
pends on the amount of damping.

Introduction
To break a wine glass, an opera singer has to sing
the right note. To hear a radio signal, you have to
be tuned to the right frequency. These are examples
of the phenomenon of resonance: a vibrating system
will respond most strongly to a force that varies with
a particular frequency.
If the motor is stopped so that the arms are locked in
place, the metal disk can still swing clockwise and
counterclockwise because it is attached to the up-
right rod with a flexible spiral spring. A push on the
disk will result in vibrations that persist for quite
a while before the internal friction in the spring re-
duces their amplitude to an imperceptible level. This
would be an example of a free vibration, in which
energy is steadily lost in the form of heat, but no
external force pumps in energy to replace it.
Suppose instead that you initially stop the disk, but
then turn on the electric motor. There is no rigid
mechanical link to the disk, since the motor and disk
are only connected through the very flexible spiral
spring. But the motor will gently tighten and loosen
the spring, resulting in the gradual building up of a
vibration in the disk.

Simplified mechanical drawing of the vibrator, front view.
Apparatus
In this lab you will investigate the phenomenon of
resonance using the apparatus shown in the figure.
1 or HP6204B 40-volt supplies, if the Leybold supplies
aren’t bought because of budget issues
Observations
A Period of Free Vibrations
Start without any of the electrical stuff hooked up.
Twist the disk to one side, release it, and determine
its period of vibration. (Both here and at points
later in the lab, you can improve your accuracy by
timing ten periods and dividing the result by ten.)
This is the natural period of the vibrations, i.e., the
period with which they occur in the absence of any
driving force.

52 Lab 19 Resonance
B Damping
Note the coils of wire at the bottom of the disk.
These are electromagnets. Their purpose is not to
attract the disk magnetically (in fact the disk is
made of a nonmagnetic metal) but rather to increase
the amount of damping in the system. Whenever a
metal is moved through a magnetic field, the elec-
trons in the metal are made to swirl around. As
they eddy like this, they undergo random collisions
with atoms, causing the atoms to vibrate. Vibration
of atoms is heat, so where did this heat energy come
from ultimately? In our system, the only source of
energy is the energy of the vibrating disk. The net
effect is thus to suck energy out of the vibration and
convert it into heat. Although this magnetic and
electrical effect is entirely different from mechanical
friction, the result is the same. Creating damping in
this manner has the advantage that it can be made
stronger or weaker simply by increasing or decreas-
ing the strength of the magnetic field.
Turn off all the electrical equipment and leave it un-
plugged. Connect the circuit shown in the top left
of the electrical diagram, consisting of a power sup-
ply to run the electromagnet plus a meter . You do
not yet need the power supply for driving the motor.
The meter will tell you how much electrical current
is flowing through the electromagnet, which will give
you a numerical measure of how strong your damp-
ing is. It reads out in units of amperes (A), the
metric unit of electrical current. Although this does
not directly tell you the amount of damping force in
units of newtons (the force depends on velocity), the
force is proportional to the current.
Once you have everything hooked up, check with
your instructor before plugging things in and turn-
ing them on. If you do the setup wrong, you could
blow a fuse, which is no big deal, but a more seri-
ous goof would be to put too much current through
the electromagnet, which could burn it up, perma-
nently ruining it. Once your instructor has checked
this part of the electrical setup she/he will show you
how to monitor the current on the meter to make
sure that you never have too much.
The Q of an oscillator is defined as the number of
oscillations required for damping to reduce the en-
ergy of the vibrations by a factor of 535 (a defini-
tion originating from the quantity e2π ). As planned
in your prelab, measure the Q of the system with
the electromagnet turned off, then with a low cur-
rent through the electromagnet, and then a higher
current. There are differences among the oscillators,
possibly because they were dropped on the floor dur-
ing the move. To compensate for this, the currents
you should use should be based on which oscillator
you have. Groups 1, 2, 3, 4, 5, 6, and 7 should use
the oscillators labeled 1, 2, 3, 4, 5, B, and D, respec-
tively. If your oscillator is 2, 3, or 5, you should use
300 mA for the low current and 600 mA for the high;
if it’s 1, 4, B, or D, use 225 and 450 mA. You will be
using these two current values throughout the lab.
C Frequency of Driven Vibration
Now connect the 24 V AC power supply to the mo-
tor.2 The motor has coarse (0 to 100) and fine (-6 to
+6) adjustment knobs. (On the older models these
are marked “groβ” and “fein,” and have different
scales.) These knobs are not calibrated in Hz, and
their readings don’t even correlate linearly with fre-
quency, so to measure the frequency of the motor
you need to use the stopwatch. However, it can be
useful to write down the knob settings that produced
a particular frequency, so that you can reproduce it
later. The older motors don’t control their frequency
as accurately as the newer ones, so if you have one
of these you should do the low-Q setup in part F.
Set the damping current to the higher of the two
values. Turn on the motor and drive the system at a
frequency very different from its natural frequency.
You will notice that it takes a certain amount of
time, perhaps a minute or two, for the system to
settle into a steady pattern of vibration. This is
called the steady-state response to the driving force
of the motor.
Does the system respond by vibrating at its natural
frequency, at the same frequency as the motor, or at
some frequency in between?
D Resonance
With your damping current still set to the higher
value, try different motor frequencies, and observe
how strong the steady-state response is. At what
motor frequency do you obtain the strongest response?
When changing from one frequency to another, you
have to allow time for the vibrator to reach its steady
state. To make it easier to tell when the steady state
is happening, it helps if you stop the vibration by
hand after changing the frequency; otherwise you
can get complicated patterns of motion in which the
exponentially decaying motion left over from the old
frequency happens on top of the new driven motion.
2 We will probably get the 24 V AC power supplies some-
time in 2009. If we don’t have those yet, use one of the power
supplies in the lab benches in 416T, setting it on 12 V and
connecting to the +12 and −12 V plugs.
53
You can save yourself some time if you think of this from the graph,
part and part F as one unit, and plan ahead so that
the data you take now are also the data you need for
part F. Analysis
E Resonance Strength Compare your observations in parts C, D, and E
with theory.
Set the motor to the resonant frequency, i.e., the
frequency at which you have found you obtain the For part F, construct graphs with the square of the
strongest response. Now measure the amplitude of amplitude on the y axis and the frequency on the
the vibrations you obtain with each of the two damp- x axis. The reason for using the square of the am-
ing currents. How does the strength of the resonance plitude is that the standard way of specifying the
depend on damping? width of a resonance peak is to give its full width at
half resonance (FWHM), which is measured between
With low amounts of damping, I have sometimes en-
the two points where the energy of the steady-state
countered a problem where the system, when driven
vibration equals half its maximum value. Energy is
near resonance, never really settles down into a steady
proportional to the square of the amplitude. Deter-
state. The amplitude varies dramatically from one
mine the FWHM of the resonance for each value of
minute to the next, perhaps because the power sup-
the damping current, and find whether the expected
ply is not stable enough to control the driving fre-
relationship exists between Q and FWHM; make a
quency consistently enough. If this happens to you,
numerical test, not just a qualitative one. Obviously
check with your instructor.
there is no way you can get an accurate FWHM if
F Width of the Resonance the peak is only as wide as a pencil on the graph —
make an appropriate choice of the range of frequen-
Now measure the response of the system for a large cies on the x axis.
number of driving frequencies, so that you can graph
the resonance curve and determine the width of the
resonance. Concentrate on the area near the top
and sides of the peak, which is what’s important for
finding the full width at half maximum (FWHM).
To make this part less time-consuming, your instruc-
tor will assign your group to do only one of the two
graphs, low-Q or high-Q. Each group will have their
own data for one Q and another group’s data for
another Q.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Plan how you will determine the Q of your os-
cillator in part B. [Hint: Note that the energy of a
vibration is proportional to the square of the ampli-
tude.]

Self-Check
Make your graphs for part F (see below), and see if
they make sense. Make sure to make the frequency
axis expanded enough to get an accurate FWHM

54 Lab 19 Resonance
55
20 Resonance (short version for physics
222)
This is a simplified version of lab 19, meant to in-
troduce some concepts related to mechanical reso-
nance, without any detailed data-taking. The idea
is to reinforce the relevant concepts from physics 221
so that they can be used as a metaphor for electrical
resonances in 222.

Apparatus
vibrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Thornton power supply . . . . . . . . . . . . . . . . . . . 1/group
stopwatch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
multimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
banana plug cables Simplified mechanical drawing of the vibrator, front view.

Goals
Observe the phenomenon of resonance.

Learn how to visualize phases and amplitudes

in a plane.

Introduction
To break a wine glass, an opera singer has to sing
the right note. To hear a radio signal, you have to
be tuned to the right frequency. These are examples
of the phenomenon of resonance: a vibrating system Electrical setup, top view.
will respond most strongly to a force that varies with
a particular frequency.
Suppose instead that you initially stop the disk, but
Apparatus then turn on the electric motor. There is no rigid
mechanical link to the disk, since the motor and disk
In this lab you will investigate the phenomenon of are only connected through the very flexible spiral
resonance using the apparatus shown in the figure. spring. But the motor will gently tighten and loosen
If the motor is stopped so that the arms are locked in the spring, resulting in the gradual building up of a
place, the metal disk can still swing clockwise and vibration in the disk.
counterclockwise because it is attached to the up-
right rod with a flexible spiral spring. A push on the
disk will result in vibrations that persist for quite Observations
a while before the internal friction in the spring re-
duces their amplitude to an imperceptible level. This A Period of Free Vibrations
would be an example of a free vibration, in which Start without any of the electrical stuff hooked up.
energy is steadily lost in the form of heat, but no Twist the disk to one side, release it, and use the
external force pumps in energy to replace it. stopwatch to determine its natural period of vibra-

56 Lab 20 Resonance (short version for physics 222)

tion. (Both here and at points later in the lab, you
can improve your accuracy by timing ten periods
and dividing the result by ten.)
B Damping
Note the coils of wire at the bottom of the disk.
These are electromagnets. Their purpose is not to
attract the disk magnetically (in fact the disk is
made of a nonmagnetic metal) but rather to increase
the amount of damping in the system. Whenever a
metal is moved through a magnetic field, the elec-
trons in the metal are made to swirl around. As
they eddy like this, they undergo random collisions
with atoms, causing the atoms to vibrate. Vibration
of atoms is heat, so where did this heat energy come
from ultimately? In our system, the only source of
energy is the energy of the vibrating disk. The net
effect is thus to suck energy out of the vibration and
convert it into heat. Although this magnetic and
electrical effect is entirely different from mechanical
friction, the result is the same. Creating damping in
this manner has the advantage that it can be made
stronger or weaker simply by increasing or decreas-
ing the strength of the magnetic field.
Turn off all the electrical equipment and leave it un-
plugged. Connect the circuit shown in the top left
of the electrical diagram, consisting of a power sup-
ply to run the electromagnet plus a meter . You do
not yet need the power supply for driving the motor.
The meter will tell you how much electrical current
is flowing through the electromagnet, which will give
you a numerical measure of how strong your damp-
ing is. It reads out in units of amperes (A), the
metric unit of electrical current. Although this does
not directly tell you the amount of damping force in
units of newtons (the force depends on velocity), the
force is proportional to the current.
Once you have everything hooked up, check with
your instructor before plugging things in and turn-
ing them on. If you do the setup wrong, you could
blow a fuse, which is no big deal, but a more seri-
ous goof would be to put too much current through
the electromagnet, which could burn it up, perma-
nently ruining it. Once your instructor has checked
this part of the electrical setup she/he will show you
how to monitor the current on the meter to make
sure that you never have too much.
The Q of an oscillator is defined as the number of
oscillations required for damping to reduce the en-
ergy of the vibrations by a factor of 535 (a definition
originating from the quantity e2π ). As planned in
your prelab, measure the Q of the system with the
electromagnet turned off, then with a current of 0.25
A through the electromagnet, and then 0.50 A. You
will be using these two current values throughout
the lab.
C Frequency of Driven Vibration
Now connect the lab’s DC power supply to the ter-
minals on the motor labeled “motorpanschluβ.” The
coarse and fine adjustments to the speed of the mo-
tor are marked “groβ” (gross) and “fein” (fine).
Set the damping current to the higher of the two
values. Turn on the motor and drive the system at a
frequency very different from its natural frequency.
You will notice that it takes a certain amount of
time, perhaps a minute or two, for the system to
settle into a steady pattern of vibration. This is
called the steady-state response to the driving force
of the motor.
Does the system respond by vibrating at its natural
frequency, at the same frequency as the motor, or at
some frequency in between?
D Resonance
With your damping current still set to the higher
value, try different motor frequencies, and observe
how strong the steady-state response is. At what
motor frequency do you obtain the strongest response?
E Resonance Strength
Set the motor to the resonant frequency, i.e., the
frequency at which you have found you obtain the
strongest response. Now measure the amplitude of
the vibrations you obtain with each of the two damp-
ing currents. How does the strength of the resonance
depend on damping?
F Phase Response
If the disk and the vertical arm were connected rigidly,
rather than through a spring, then they would al-
ways be in phase. For instance, the disk would reach
its most extreme clockwise angle at the same mo-
ment when the vertical arm was also all the way
clockwise. But since the connection is not rigid,
this need not be the case. Find a frequency sig-
nificantly below the resonant frequency, at which
the amplitude of the steady-state response is per-
haps one tenth of the value it would have at res-
onance. What do you observe about the relative
phase of the disk and the vertical arm? Are they in
phase or out of phase? You can describe the phase
by assigning positive phase angles to oscillations in
which the disk is ahead of the arm, and negative
phases when the disk is behind. These phase angles
can range from -180 ◦ to 180 ◦ . Actually +180 ◦ and
57
-180 ◦ would represent the same thing: the oscilla- tude.]
tions have phases that are exactly the opposite. Try
to estimate roughly what the phase angle is. You
don’t have any way to measure it accurately, but you
should be able to estimate it to the nearest multiple
of 45 ◦ . Measure the amplitude of the steady-state
response as well.
Now measure the phase and amplitude of the re-
sponse when the driving force is at the resonant fre-
quency.
Finally, do the same measurements when the driving
force is significantly above resonance.

Analysis
The point of this is to connect the mechanical analog
to what you know about the phase response of a
resonant LRC circuit. You’re measuring the phase
between F and x, which is analogous to the phase
between V and q in electrical terms. However, most
people think of AC circuits in terms of V and I, not
V and q. The phase relationships you’re expecting,
therefore, are those that would hold between F and
v = dx/dt, which differ by 90 degrees from the F − x
phases you actually measured as raw data.
To complete the electrical analogy, we would really
prefer to discuss the mechanical analog of impedance.
The (constant) driving force from the motor plays
the role of the voltage, while the frequency-dependent
amplitude of the vibration plays the role of the cur-
rent. Dividing these two quantities gives us some-
thing analogous to impedance, and since the driving
force is always the same, we can say that the in-
verse of the amplitude is essentially a measure of
the impedance.
To summarize, you have a complex impedance whose
amplitude and phase angle you can determine from
your data. Plot the impedances at the various fre-
quencies in the complex plane.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Plan how you will determine the Q of your os-
cillator in part B. [Hint: Note that the energy of a
vibration is proportional to the square of the ampli-

58 Lab 20 Resonance (short version for physics 222)

59
21 Standing Waves
Apparatus
string
weights, including 1-gram weights
pulley
vibrator
paperclips
metersticks
butcher paper
scissors
weight holders
Goals
Observe the resonant modes of vibration of a
string.
Find how the speed of waves on a string de-
pends on the tension in the string.
Introduction
The Greek philosopher Pythagoras is said to have
been the first to observe that two plucked strings
sounded good together when their lengths were in
the proportion of two small integers. (This is assum-
ing the strings are of the same material and under
the same tension.) For instance, he thought a pleas-
ant combination of notes was produced when one
string was twice the length of the other, but that the
combination was unpleasant when the ratio was, say,
1.4 to 1 (like the notes B and F). Although different
combinations of notes are used in different cultures
and different styles of music, there is at least some
scientific justification for Pythagoras’ statement. We
now know that a plucked string does not just vibrate
at a single frequency but simultaneously at a whole
series of frequencies f1 , 2f1 , 3f1 ,... These frequen-
cies are called the harmonics. If one string is twice
the length of the other, then its lowest harmonic is at
half the frequency of the other string’s, and its har-
monics coincide with the odd-numbered harmonics
of the other string. If the ratio is 1.4 to 1, however,
then there is essentially no regular relationship be-
tween the two sets of frequencies, and many of the
harmonics lie close enough in frequency to produce
unpleasant beats.
60 Lab 21 Standing Waves
Setup
The apparatus allows you to excite vibrations at a
fixed frequency of 120 Hz (twice the frequency of
the alternating current that runs the vibrator). The
tension in the string can be controlled by varying the
weight.
You may find it helpful to put a strip of white butcher
paper behind the black string for better visual con-
trast.
It’s important to get the vibrator set up properly
along the same line as the string, not at an angle.
Observations
Observe as many modes of vibration as you can. You
will probably not be able to observe the fundamen-
tal (one hump) because it would require too much
weight. In each case, you will want to fine-tune the
weight to get as close as possible to the middle of
the resonance, where the amplitude of vibration is
at a maximum. When you’re close to the peak of
a resonance, an easy way to tell whether to add or
remove weight is by gently pressing down or lifting
up on the weights with your finger to see whether
the amplitude increases or decreases.
For large values of N , you may find that you need to
use a paperclip instead of the weight holder, in order
to make the mass sufficiently small. Keep in mind,
however, that you won’t really improve the quality
of your data very much by taking data for very high
values of N , since the 1-gram precision with which
you can locate these resonances results in a poor
relative precision compared to a small weight.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise ity of the waves in the string and the tension in the
you’re just setting yourself up for failure in lab. string. (Do not just try to find the correct power
law in the textbook, because besides observing the
P1 Should the whole length of the string be counted
phenomenon of resonance, the point of the lab is to
in L, or just the part from the vibrator to the pulley?
prove experimentally what the power-law relation-
P2 How is the tension in the string, T , related to ship is.)
the mass of the hanging weight?
P3 The figure below shows the N = 1, 2, and 3
patterns of vibration. Suppose the length of the
string is one meter. In each case, find the wave-
length.

P4 Generalize your numerical results from P3 to

give a general equation for λ in terms of N and L,
the length of the string. Check its units, and check
that it recovers the special cases done numerically in
P3.
P5 How can the velocity of the waves be deter-
mined if you know the frequency, f , the length of
the string, L, and the number of humps, N ?

Self-Check
Do your analysis in lab.

Analysis
Use the techniques given in appendix 5 to see if you
can find a power-law relationship between the veloc-

61
22 Resonances of Sound
Apparatus
wave generator (PASCO PI-9587C) . . . . . . . 1/group
speaker (Thornton) . . . . . . . . . . . . . . . . . . . . . . . 1/group
100 mL graduated cylinder . . . . . . . . . . . . . . . 1/group
Linux computers with FFT Explorer installed (in
416 and 416P)
flexible whistling tube . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
tuning fork marked with frequency, mounted on a
wooden box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
aluminum rod, 3/4-inch dia, about 1 m long
2 that, which is why electronic synthesizers still do not
Goals
Find the resonant frequencies of the air inside
a cylinder by two methods.
Measure the speeds of sound in air and in alu-
minum.
Introduction
In the womb, your first sensory experiences were of
your mother’s voice, and soon after birth you learned
to distinguish the particular sounds of your parents’
voices from those of strangers. The human ear-brain
system is amazingly sophisticated in its ability to
classify vowels and consonants, recognize people’s
voices, and analyze musical sound. Until the 19th-
century investigations of Helmholtz, the whole pro-
cess was completely mysterious. How could we so
easily tell a cello from a violin playing the same note?
A radio station in Chicago has a weekly contest in
which jazz fanatics are asked to identify instrumen-
talists simply by their distinctly individual timbres
— how is this possible?
Helmholtz found (using incredibly primitive nonelec-
tronic equipment) that part of the answer lay in the
relative strengths of the overtones. The psychologi-
cal sensation of pitch is related to frequency, e.g., 440
Hz is the note “A.” But a saxophonist playing the
note “A” is actually producing a rich spectrum of
frequencies, including 440 Hz, 880 Hz, 1320 Hz, and
many other multiples of the lowest frequency, known
as the fundamental. The ear-brain system perceives
all these overtones as a single sound because they are
all multiples of the fundamental frequency. (The Ja-
vanese orchestra called the gamelan sounds strange
62 Lab 22 Resonances of Sound
to westerners partly because the various gongs and
cymbals have overtones that are not integer multi-
ples of the fundamental.)
One of the things that would make “A” on a clarinet
sound different from “A” on a saxophone is that the
880 Hz overtone would be quite strong for the sax-
ophone, but almost entirely missing for the clarinet.
Although Helmholtz thought the relative strengths
of the overtones was the whole story when it came
to musical timbre, actually it is more complex than
sound as good as acoustic instruments. The timbre
depends not just on the general strength of the over-
tones but on the details of how they first build up
(the attack) and how the various overtones fade in
and out slightly as the note continues.
Why do different instruments have different sound
spectra, and why, for instance, does a saxophone
have an overtone that the clarinet lacks? Many mu-
sical instruments can be analyzed physically as tubes
that have either two open ends, two closed ends, or
one open end and one closed end. The overtones
correspond to specific resonances of the air column
inside the tube. A complete treatment of the subject
is given in your textbook, but the basic principle is
that the resonant standing waves in the tube must
have an antinode (point of maximum vibration) at
any closed end of the tube, and a node (point of zero
vibration) at any open end.
Using the Wave Generator
The wave generator works like the amplifier of your
stereo, but instead of playing a CD, it produces a
sine wave whose frequency and amplitude you can
control. By connecting it to a speaker, you can con-
vert its electrical currents to sound waves, making
a pure tone. The frequency of the sine wave cor-
responds to musical pitch, and the amplitude cor-
responds to loudness. Use the output labeled LO
Ω. The wave generator can create enough voltage to
give a mildly unpleasant tingling sensation in your
hand if you touch the leads. None of the electri-
cal apparatus used in this lab, however, is any more
dangerous than a home TV or stereo.
Setup
Unplug the wave generator. Check the fuse in the
back of the wave generator to make sure it is not
blown, then put it back in. Plug in the wave gener-
ator and turn off the on/off switch at the top right.
Turn the “amplitude” knob of the wave generator to
zero, and then turn on the on/off switch.
Plug the speaker into the wave generator. The ba-
nana plugs go in the two holes on the right. Set the
frequency to something audible. Wait 30 seconds
for the wave generator to warm up, then turn the
amplitude knob up until you hear a sound.
The wave generator and the speaker are not really
designed to work together, so if you leave the volume
up very high for a long time, it is possible to blow
the speaker or damage the wave generator. Also, the
sine waves are annoying when played continuously at
loud volumes!
Preliminary Observations
Observations
This lab has three parts, A, B, and C. It is not really
possible for more than one group to do part A in the
same room, both because their sounds interfere with
one another and because the noise becomes annoy-
ing for everyone. Your instructor will probably have
three groups working on part A at one time, one
group in the main room, one in the small side room,
and one in the physics stockroom. Meanwhile, the
other groups will be doing parts B and C.
A Direct Measurement of Resonances by Lis-
tening
Set up the graduated cylinder so its mouth is cov-
ering the center of the speaker. Find as many fre-
quencies as possible at which the cylinder resonates.
When you sweep through those frequencies, the sound
becomes louder. To make sure you’re really hearing
a resonance of the cylinder, make sure to repeat each
observation with the cylinder removed, and make
sure the resonance goes away. For each resonance,
take several measurements of its frequency — if you
are careful, you can pin it down to within ±10 Hz
or so. You can probably speed up your search sig-
nificantly by calculating approximately where you
expect the resonances to be, then looking for them.
B Electronic Measurement of Resonances of
an Air Column
The resonances of the air column in a cylinder can
also be excited by a stream of air flowing over an
opening, as with a flute. In this part of the lab, you
will excite resonances of a long, flexible plastic tube
by grabbing it at one end and swinging it in a cir-
cle. The frequency of the sound will be determined
electronically. Note that your analysis for these res-
onances will be somewhat different, since the tube
is open at both ends, and it therefore has different
patterns of resonances from the graduated cylinder,
which was only open at one end.
To measure the frequency, you will use a computer to
analyze the sound. There are two Linux computers
that have the right software and hardware. As a
warmup before attempting the actual measurements
with the whistling tube, try the following. First,
start up the program if nobody else has already done
so. It is called FFT Explorer, and you can run it by
double-clicking on its icon on the desktop. In real
time, the program will monitor the sound coming
into the microphone, and display a graph of loudness
versus frequency. Try whistling. The frequency at
which you whistled should show up as a prominent
peak. You may need to play with the frequency and
loudness scales, using the two menus on the lower
right. If you’re not careful, it’s easy to get confused
by setting a frequency range that’s too narrow, so
that the peak you want isn’t even on the graph. It’s
a good idea to try it first on a very wide frequency
scale, and then narrow the scale to the narrowest one
that allows you to see the peak. When you get the
graph you want to see, you can freeze it by clicking
on the stop button. Although the software doesn’t
give any convenient way to read off the frequency
of the peak with high precision, you can accomplish
that by measuring on the screen with a ruler, and
interpolating.
Debugging software problems:
If sound input doesn’t work, or mysteriously
stops working, it’s typically because Linux’s
sound system (called ALSA) is upset; this can
be fixed by logging out, and then logging back
in again.
Now try the whole procedure with the tuning fork
instead of whistling, and make sure you can use the
computer to obtain the frequency inscribed on the
fork. You can put the mic inside the wooden box
that the tuning fork is mounted on. Although the
software doesn’t let yu zoom in on the peak, you
63
can lay a ruler on the screen, and interpolate fairly
accurately.
Once you have done these warmups, you are ready to
analyze the sound from the whistling tube. You only
need to analyze data from one frequency, although
if you’re not sure which mode you produced, it may
be helpful to observe the pattern of the frequencies.
(If you guess wrong about which mode it was, you’ll
find out, because the value you extract for the speed
of sound will be way off.)
C The Speed of Sound in Aluminum
The speed of sound in dense solid is much faster
than its speed in air. In this part of the lab, you
will extract the speed of sound in aluminum from
a measurement of the lowest resonant frequency of
a solid aluminum rod. You will use the computer
for an electronic measurement of the frequency, as
in part B.
Grab the rod with two fingers exactly in the middle,
hold it vertically, and tap it on the floor. You will
hear two different notes sounding simultaneously. A
quick look at their frequencies shows that they are
not in a 2:1 ratio as we would expect based on our
experiences with symmetric wave patterns. This is
because these two frequencies in the rod are actu-
ally two different types of waves. The higher note is
produced by longitudinal compression waves, which
means that an individual atom of aluminum is mov-
ing up and down the length of the rod. This type
of wave is analogous to sound waves in air, which
are also longitudinal compression waves. The lower
note comes from transverse vibrations, like a vibrat-
ing guitar string. In the transverse vibrations, atoms
are moving from side to side, and the rod as a whole
is bending.
If you listen carefully, you can tell that the trans-
verse vibration (the lower note) dies out quickly, but
the longitudinal mode keeps going for a long time.
That gives you an easy way to isolate the longitudi-
nal mode, which is the one we’re interested in; just
wait for the transverse wave to die out before you
freeze the graph on the computer.
The rod is symmetric, so we expect its longitudi-
nal wave patterns to be symmetric, like those of the
whistling tube. The rod is different, however, be-
cause whereas we can excite a variety of wave pat-
terns in the tube by spinning it at different speeds,
we find we only ever get one frequency from the rod
by tapping it at its end: it appears that there is
only one logitudinal wave pattern that can be ex-
cited strongly in the rod by this method. The prob-
64 Lab 22 Resonances of Sound
lem is that we then need to infer what the pattern is.
Since you hold the rod at its center, friction should
very rapidly damp out any mode of vibration that
has any motion at the center. Therefore there must
be a node at the center. We also know that at the
ends, the rod has nothing to interact with but the
air, and therefore there is essentially no way for any
significant amount of wave energy to leak out; we
therefore expect that waves reaching the ends have
100% of their energy reflected. Since energy is pro-
portional to the square of amplitude, this means
that a wave with unit amplitude can be reflected
from the ends with an amplitude of either R = +1
(100% uninverted reflection) or −1 (100% inverted).
In the R = −1 case, the reflected wave would can-
cel out the incident wave at the end of the rod, and
we would have a node at the end, as in lab 21. In
the R = +1 case, there would be an antinode. But
when you tap the end of the rod on the floor, you are
evidently exciting wave motion by moving the end,
and it would not be possible to excite vibrations by
this method if the vibrations had no motion at the
end. We therefore conclude that the rod’s pattern
of vibration must have a node at the center, and
antinode at the ends. There is an infinite number of
possible wave patterns of this kind, but we will as-
sume that the pattern that is excited strongly is the
one with the longest wavelength, i.e., the only node
is at the center, and the only antinodes are those at
the ends. If you feel like it, there are a couple of
possible tests you can try do to check whether this
is the right interpretation. One is to see if you can
detect any other frequencies of longitudinal vibra-
tion that are excited weakly. Another is to predict
where the other nodes would be, if there were more
than one, and then see if the vibration is killed by
touching the rod there with your other hand; if there
is a node there, touching it should have no effect.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Find an equation to predict the frequencies of
the resonances in parts A and B. Note that they
will not be the same equations, since one tube is
symmetric and the other is asymmetric.
Self-Check
Extract the speed of sound from either part A or
part B, without error analysis, and make sure you
get something close to the accepted value.

Analysis
Make a graph of wavelength versus period for the
resonances of the graduated cylinder, check whether
it looks like it theoretically should, and if so, find
the speed of sound from its slope, with error bars,
as discussed in appendix 4.
Use the data from part B to find a second value of
the speed of sound, also with error bars.
The effective length of the cylinder in part A should
be increased by 0.4 times its diameter to account for
the small amount of air beyond the end that also vi-
brates. For part B, where the whistling tube is open
at both ends, you should add 0.8 times its diameter.
When estimating error bars from part B, you may
be tempted to say that it must be perfectly accurate,
since its being done by a computer. Not so! You will
see that the peak is a little ragged, and that means
you cannot find the frequency with perfect accuracy.
Extract the speed of sound in aluminum from your
data in part C, including error bars.

65
23 Static Electricity
Apparatus
scotch tape
rubber rod
heat lamp
fur
bits of paper
rods and strips of various materials
30-50 cm rods, and angle brackets, for hanging charged
rods
Goal
Determine the qualitative rules governing electrical
charge and forces.
Introduction
Newton’s law of gravity gave a mathematical for-
mula for the gravitational force, but his theory also
made several important non-mathematical statements
about gravity:
Every mass in the universe attracts every other
mass in the universe.
Gravity works the same for earthly objects as
for heavenly bodies.
The force acts at a distance, without any need
for physical contact.
Mass is always positive, and gravity is always
attractive, not repulsive.
The last statement is interesting, especially because
it would be fun and useful to have access to some
negative mass, which would fall up instead of down
(like the “upsydaisium” of Rocky and Bullwinkle
fame).
Although it has never been found, there is no theo-
retical reason why a second, negative type of mass
can’t exist. Indeed, it is believed that the nuclear
force, which holds quarks together to form protons
and neutrons, involves three qualities analogous to
mass. These are facetiously referred to as “red,”
“green,” and “blue,” although they have nothing to
do with the actual colors. The force between two of
66 Lab 23 Static Electricity
the same “colors” is repulsive: red repels red, green
repels green, and blue repels blue. The force be-
tween two different “colors” is attractive: red and
green attract each other, as do green and blue, and
red and blue.
When your freshly laundered socks cling together,
that is an example of an electrical force. If the grav-
itational force involves one type of mass, and the
nuclear force involves three colors, how many types
of electrical “stuff” are there? In the days of Ben-
jamin Franklin, some scientists thought there were
two types of electrical “charge” or “fluid,” while oth-
ers thought there was only a single type. In this lab,
you will try to find out experimentally how many
types of electrical charge there are.
Observations
Stick a piece of scotch tape on a table, and then lay
another piece on top of it. Pull both pieces off the
table, and then separate them. If you now bring
them close together, you will observe them exerting
a force on each other. Electrical effects can also be
created by rubbing the fur against the rubber rod.
Your job in this lab is to use these techniques to
test various hypotheses about electric charge. The
most common difficulty students encounter is that
the charge tends to leak off, especially if the weather
is humid. If you have charged an object up, you
should not wait any longer than necessary before
making your measurements. It helps if you keep your
hands dry.
A Repulsion and/or attraction
Test the following hypotheses. Note that they are
mutually exclusive, i.e., only one of them can be true.
A1) Electrical forces are always attractive.
A2) Electrical forces are always repulsive.
A3) Electrical forces are sometimes attractive and
sometimes repulsive.
Interpretation: Once you think you have tested these
hypotheses fairly well, discuss with your instructor
what this implies about how many different types of
charge there might be.
B Are there forces on objects that have not been
specially prepared?
So far, special preparations have been necessary in
order to get objects to exhibit electrical forces. These
preparations involved either rubbing objects against
each other (against resistance from friction) or pulling
objects apart (e.g. overcoming the sticky force that
holds the tape together). In everyday life, we do not
seem to notice electrical forces in objects that have
not been prepared this way.
Now try to test the following hypotheses. Bits of pa-
per are a good thing to use as unprepared objects,
since they are light and therefore would be easily
moved by any force. Do not use tape as an un-
charged object, since it can become charged a little
bit just by pulling off the roll.
B1) Objects that have not been specially prepared
are immune to electrical forces.
B2) Unprepared objects can participate in electrical
forces with prepared objects, and the forces involved
are always attractive.
B3) Unprepared objects can participate in electrical
forces with prepared objects, and the forces involved
are always repulsive.
B4) Unprepared objects can participate in electrical
forces with prepared objects, and the forces involved
can be either repulsive of attractive.
Hypotheses B1 through B4 are mutually exclusive.
C Rules of repulsion and/or attraction and the
number of types of charge
Test the following mutually exclusive hypotheses:
C1) There is only one type of electric charge, and
the force is always attractive.
C2) There is only one type of electric charge, and
the force is always repulsive.
C3) There are two types of electric charge, call them
X and Y. Like charges repel (X repels X and Y repels
Y) and opposite charges attract (X and Y attract
each other).
C4) There are two types of electric charge. Like
charges attract and opposite charges repel.
C5) There are three types of electric charge, X, Y
and Z. Like charges repel and unlike charges attract.
The only way to keep all your observations straight
is to make a square table, in which the rows and
columns correspond to the different objects you’re
testing against each other for attraction and repul-
sion. To test C3 versus C5, you’ll need to see if you
can successfully explain your whole table by labeling
the objects with only two labels, X and Y.
Some of the equipment may look identical, but not
be identical. In particular, some of the clear rods
have higher density than others, which may be be-
cause they’re made different types of plastic, or glass.
This could affect your conclusions, so you may want
to check, for example, whether two rods with the
same diameter, that you think are made of the same
material, actually weigh the same.
In general, you will find that some materials, and
some combinations of materials, are more easily charged
than others. For example, if you find that the ma-
hogony rod rubbed with the weasel fur doesn’t charge
well, then don’t keep using use it! The white plastic
strips tend to work well, so don’t neglect them.
Discuss your conclusions with your instructor.
Self-Check
The following are examples of incorrect reasoning
about this lab. As a self-check, it would be a very
good idea to figure out for yourself in each case why
the reasoning is logically incorrect or inconsistent
with Newton’s laws. You do not need to do this in
writing — it is just to help you understand what’s
going on. If you can’t figure some of them out, ask
your instructor before leaving lab.
(1) “The first piece of tape exerted a force on the
second, but the second didn’t exert one on the first.”
(2) “The first piece of tape repelled the second, and
the second attracted the first.”
(3) “We observed three types of charge: two that
exert forces, and a third, neutral type.”
(4) “The piece of tape that came from the top was
positive, and the piece from the bottom was nega-
tive.”
(5) “One piece of tape had electrons on it, and the
other had protons on it.”
(6) “We know there were two types of charge, not
three, because we observed two types of interactions,
attraction and repulsion.”
Writeup
Explain what you have concluded about electrical
charge and forces. Base your conclusions on your
data!
67
Notes For Next Week
(1) Next week, when you turn in your writeup for
this lab, you also need to turn in a prelab writeup
for the next lab. The prelab questions are listed
at the end of the description of that lab in the lab
manual. Never start a lab without understanding
the answers to all the prelab questions; if you turn
in partial answers or answers you’re unsure of, dis-
cuss the questions with your instructor or with other
students to make sure you understand what’s going
on.
(2) You should exchange phone numbers with your
lab partners for general convenience throughout the
semester. You can also get each other’s e-mail ad-
dresses by logging in to Spotter and clicking on “e-
mail.”

Rules and Organization for

Professor Crowell’s Labs
Collection of raw data is work you share with your
lab partners. Once you’re done collecting data, you
need to do your own analysis. E.g., it is not okay for
two people to turn in the same calculations, or on a
lab requiring a graph for the whole group to make
one graph and turn in copies.
As explained in the syllabus, you’ll do some labs as
formal writeups, others as informal “check-off” labs.
The format of formal lab writeups is given in ap-
pendix 1 on page 161. The raw data section must
be contained in your bound lab notebook. Typically
people word-process the abstract section, and any
other sections that don’t include much math, and
stick the printout in the notebook to turn it in. The
justification and reasoning section will usually just
consist of hand-written calculations you do in your
lab notebook. You need two lab notebooks, because
on days when you turn one in, you need your other
one to take raw data in for the next lab. You may
find it convenient to leave one or both of your note-
books in the cupboard at your lab bench whenever
you don’t need to have them at home to work on;
this eliminates the problem of forgetting to bring
your notebook to school.
For a check-off lab, the main thing I’ll pay attention
to is your abstract. The rest of your work for a
check-off lab can be informal, and I may not ask to
see it unless I think there’s a problem after reading
your abstract.

68 Lab 23 Static Electricity

69
24 The Oscilloscope
Apparatus your lips, the flowing air is like an electrical current,
and the difference in pressure between your mouth
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group and the room is like the difference in voltage. For
microphone (RS 33-1067) . . . . . . . . . . . . . . . . . 1/group the purposes of this lab, it is not really necessary
microphone (Shure C606) . . . . . . . . . . . . . . . . . . . . . . . .1 for you to work with the fundamental definition of
PI-9587C sine wave generator . . . . . . . . . . . . . 1/group voltage.
amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
various tuning forks The input connector on the front of the oscilloscope
accepts a type of cable known as a BNC cable. A
BNC cable is a specific example of coaxial cable
If there’s an equipment conflict with respect to the (“coax”), which is also used in cable TV, radio, and
sine wave generators, the HP200CD sine wave gen- computer networks. The electric current flows in
erators can be used instead. one direction through the central conductor, and re-
turns in the opposite direction through the outside
conductor, completing the circuit. The outside con-
Goals ductor is normally kept at ground, and also serves as
Learn to use an oscilloscope. shielding against radio interference. The advantage
of coaxial cable is that it is capable of transmitting
Observe sound waves on an oscilloscope. rapidly varying signals without distortion.

Introduction
One of the main differences you will notice between
your second semester of physics and the first is that
many of the phenomena you will learn about are
not directly accessible to your senses. For example,
electric fields, the flow of electrons in wires, and the
inner workings of the atom are all invisible. The
oscilloscope is a versatile laboratory instrument that Most of the voltages we wish to measure are not big
can indirectly help you to see what’s going on. enough to use directly for the vertical deflection volt-
age, so the oscilloscope actually amplifies the input
The Oscilloscope voltage, i.e., the small input voltage is used to con-
trol a much large voltage generated internally. The
amount of amplification is controlled with a knob on
An oscilloscope graphs an electrical signal that varies the front of the scope. For instance, setting the knob
as a function of time. The graph is drawn from left to on 1 mV selects an amplification such that 1 mV at
right across the screen, being painted in real time as the input deflects the electron beam by one square
the input signal varies. In this lab, you will be using of the 1-cm grid. Each 1-cm division is referred to
the signal from a microphone as an input, allowing as a “division.”
you to see sound waves.
The input signal is supplied in the form of a voltage. The Time Base and Triggering
You are already familiar with the term “voltage”
from common speech, but you may not have learned
Since the X axis represents time, there also has to
the formal definition yet in the lecture course. Volt-
be a way to control the time scale, i.e., how fast
age, measured in metric units of volts (V), is defined
the imaginary ”penpoint” sweeps across the screen.
as the electrical potential energy per unit charge.
For instance, setting the knob on 10 ms causes it to
For instance if 2 nC of charge flows from one ter-
sweep across one square in 10 ms. This is known as
minal of a 9-volt battery to the other terminal, the
the time base.
potential energy consumed equals 18 nJ. To use a
mechanical analogy, when you blow air out between In the figure, suppose the time base is 10 ms. The

70 Lab 24 The Oscilloscope


scope has 10 divisions, so the total time required for
the beam to sweep from left to right would be 100
ms. This is far too short a time to allow the user
to examine the graph. The oscilloscope has a built-
in method of overcoming this problem, which works
well for periodic (repeating) signals. The amount
of time required for a periodic signal to perform its
pattern once is called the period. With a periodic
signal, all you really care about seeing what one pe-
riod or a few periods in a row look like — once you’ve
seen one, you’ve seen them all. The scope displays
one screenful of the signal, and then keeps on over-
laying more and more copies of the wave on top of
the original one. Each trace is erased when the next
one starts, but is being overwritten continually by
later, identical copies of the wave form. You simply
see one persistent trace.
How does the scope know when to start a new trace?
If the time for one sweep across the screen just hap-
pened to be exactly equal to, say, four periods of the
signal, there would be no problem. But this is un-
likely to happen in real life — normally the second
trace would start from a different point in the wave-
form, producing an offset copy of the wave. Thou-
sands of traces per second would be superimposed
on the screen, each shifted horizontally by a differ-
ent amount, and you would only see a blurry band
of light.
To make sure that each trace starts from the same
point in the waveform, the scope has a triggering cir-
cuit. You use a knob to set a certain voltage level,
the trigger level, at which you want to start each
trace. The scope waits for the input to move across
the trigger level, and then begins a trace. Once that
trace is complete, it pauses until the input crosses
the trigger level again. To make extra sure that it is
really starting over again from the same point in the
waveform, you can also specify whether you want to
start on an increasing voltage or a decreasing volt-
age — otherwise there would always be at least two
points in a period where the voltage crossed your
trigger level.
Setup
To start with, we’ll use a sine wave generator, which
makes a voltage that varies sinusoidally with time.
This gives you a convenient signal to work with while
you get the scope working. Use the black and white
outputs on the PI-9587C.
The figure on the last page is a simplified drawing
of the front panel of a digital oscilloscope, showing
only the most important controls you’ll need for this
lab. When you turn on the oscilloscope, it will take
a while to start up.
Preliminaries:
Press DEFAULT SETUP.
Use the SEC/DIV knob to put the time base
on something reasonable compared to the pe-
riod of the signal you’re looking at. The time
base is displayed on the screen, e.g., 10 ms/div,
or 1 s/div.
Use the VOLTS/DIV knob to put the voltage
scale (Y axis) on a reasonable scale compared
to the amplitude of the signal you’re looking
at.
The scope has two channels, i.e., it can ac-
cept input through two BNC connectors and
display both or either. You’ll only be using
channel 1, which is the only one represented in
the simplified drawing. By default, the oscil-
loscope draws graphs of both channels’ inputs;
to get rid of ch. 2, hold down the CH 2 MENU
button (not shown in the diagram) for a couple
of seconds. You also want to make sure that
the scope is triggering on CH 1, rather than
CH 2. To do that, press the TRIG MENU
button, and use an option button to select CH
1 as the source. Set the triggering mode to
normal, which is the mode in which the trig-
gering works as I’ve described above. If the
trigger level is set to a level that the signal
never actually reaches, you can play with the
knob that sets the trigger level until you get
something. A quick and easy way to do this
without trial and error is to use the SET TO
71
 50automatically sets the trigger level to mid-
way between the top and bottom peaks of the
signal.
You want to select AC, not DC or GND, on
the channel you’re using. You are looking at
a voltage that is alternating, creating an al-
ternating current, “AC.” The “DC” setting is
only necessary when dealing with constant or
very slowly varying voltages. The “GND” sim-
ply draws a graph using y = 0, which is only
useful in certain situations, such as when you
can’t find the trace. To select AC, press the
CH 1 MENU button, and select AC coupling.
Observe the effect of changing the voltage scale and
time base on the scope. Try changing the frequency
and amplitude on the sine wave generator.
You can freeze the display by pressing RUN/STOP,
and then unfreeze it by pressing the button again.
Preliminary Observations
Now try observing signals from the microphone. By
feeding the mic’s signal through the amplifier and
then to the scope, you can make the signals easier
to see.
As of fall 2008, we’re in the process of testing a bet-
ter mic (Shure brand) to replace the Radio Shack
ones. We have one of the Shure ones. If your group
is the one that gets it, please relay the information
about how it worked through your instructor and
back to Ben Crowell. Some notes about this mic: As
with the Radio Shack mics, polarity matters. The
tip of the phono plug connector is the live connec-
tion, and the part farther back from the tip is the
grounded part. You can connect on to the phono
plug with alligator clips. You don’t need the am-
plifier. Notes for instructors: This mic was $30 at
Fry’s. It has an unusually high gain, -52 dBV/Pa at
1 kHz, which helps to make the signals clean enough
to see well on a scope without preamplification. Its
output impedance is 600 ohms. The main reason the
RS 33-1067 mics have a poorer S/N ratio is that the
cables are not coax, so they pick up a lot of noise
in differential mode. The RS 33-3013 mics are not
really any better for this application; although they
do have coax cables, they have a very low gain. We
should buy phono-to-BNC connectors for the Shure
mics.
Once you have your setup working, try measuring
the period and frequency of the sound from a tuning
fork, and make sure your result for the frequency is
72 Lab 24 The Oscilloscope
the same as what’s written on the tuning fork.
Don’t crank the gain on the amplifier all the way
up. If you do, the amplifier will put out a distorted
waveform. Use the highest gain you can use without
causing distortion.
Observations
A Periodic and nonperiodic speech sounds
Try making various speech sounds that you can sus-
tain continuously: vowels or certain consonants such
as “sh,” “r,” “f” and so on. Which are periodic and
which are not?
Note that the names we give to the letters of the
alphabet in English are not the same as the speech
sounds represented by the letter. For instance, the
English name for “f” is “ef,” which contains a vowel,
“e,” and a consonant, “f.” We are interested in the
basic speech sounds, not the names of the letters.
Also, a single letter is often used in the English writ-
ing system to represent two sounds. For example,
the word “I” really has two vowels in it, “aaah” plus
“eee.”
B Loud and soft
What differentiates a loud “aaah” sound from a soft
one?
C High and low pitch
Try singing a vowel, and then singing a higher note
with the same vowel. What changes?
D Differences among vowel sounds
What differentiates the different vowel sounds?
E Lowest and highest notes you can sing
What is the lowest frequency you can sing, and what
is the highest?
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 In the sample oscilloscope trace shown on page
70, what is the period of the waveform? What is its
frequency? The time base is 10 ms.
P2 In the same example, again assume the time
base is 10 ms/division. The voltage scale is 2 mV/division.
Assume the zero voltage level is at the middle of
the vertical scale. (The whole graph can actually be
shifted up and down using a knob called “position.”)
What is the trigger level currently set to? If the trig-
ger level was changed to 2 mV, what would happen
to the trace?
P3 Referring to the chapter of your textbook on
sound, which of the following would be a reasonable
time base to use for an audio-frequency signal? 10
ns, 1µ s, 1 ms, 1 s
P4 Does the oscilloscope show you the period of
the signal, or the wavelength? Explain. (If you’re in
Physics 222, skip this one, because you don’t know
about the definition of wavelength yet.)

Analysis
The format of the lab writeup can be informal. Just
describe clearly what you observed and concluded.

73
A simplified diagram of the controls on a digital oscilloscope.

74 Lab 24 The Oscilloscope

75
25 The Speed of Sound
Based on a lab by Hans Rau.
Apparatus
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
optical bench . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
HP function generator . . . . . . . . . . . . . . . . . . . . 1/group
transducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
Goal
Measure the speed of sound.
Introduction
There are several simple methods for getting a rough
estimate of the speed of sound, for instance tim-
ing an echo, or watching the kettledrum player at
a symphony and seeing how long the sound takes to
arrive after you see the mallet strike the drumhead.
The latter method, comparing vision against hear-
ing, assumes that the speed of light is much greater
than the speed of sound, the same assumption that
is used when estimating the distance to a lightning
strike based on the interval between the flash and
the thunder. The assumption is a good one, since
light travels about a million times faster than sound.
Military jets routinely exceed the speed of sound,
but no human has ever traveled at speeds even re-
motely comparable to that of light. (The electrons
in your television set are moving at a few percent of
the speed of light, and velocities of 0.999999999999
times the speed of light can be attained in particle
accelerators. According to Einstein’s theory of rela-
tivity, motion faster than light is impossible.)
In this lab, you will make an accurate measurement
of the speed of sound by measuring the wavelength
and frequency of a pure tone (sine wave) and com-
puting
v = λf .
We will be using sound with a frequency of about
35-40 kHz, which is too high to be audible. This has
the advantage of eliminating the annoying din of six
lab groups producing sine waves at once. Such high-
frequency, inaudible sound is known as ultrasound.
Ultrasound at even higher frequencies, in the MHz
76 Lab 25 The Speed of Sound
range, is used for imaging fetuses in the womb.
Setup
The setup is shown below. A transducer is a de-
vice capable of acting as either a speaker or a micro-
phone. The function generator is used to create a
voltage that varies sinusoidally over time. This volt-
age is connected through two coax cables, to the os-
cilloscope and the first transducer, used as a speaker.
The sound waves travel from the first transducer to
the second transducer, used as a microphone. You
will be using both channels of the scope to display
graphs of two waveforms at the same time on the
oscilloscope. As you slide one transducer along the
optical bench, changing the distance between them,
you will change the phase of one wave relative to the
other. Thus, you can determine the distance corre-
sponding to a given number of wavelengths and ex-
tract the wavelength of the sound waves accurately.
The wavelength of the sound will be roughly a few
cm. The frequency can be read from the knob on
the function generator. (The time scale of an os-
cilloscope typically has a systematic error of about
2-5%, so you should not use a measurement of the
period from the scope for this purpose.)
When setting up the scope, you will need to select
one channel or the other to trigger on. You can
select the voltage scales for the two channels inde-
pendently, but they always have the same time base.
The most common problem in this lab is that some
electrical current gets through the metal optical bench,
causing the receiving transducer to pick up the orig-
inal input signal directly, rather than by receiving
the sound waves. A precaution that usually works
is to connect the optical bench to the ground con-
tact of the scope (use an alligator clip to attach to
the body of the bench). It is easy to check whether
the problem exists: put your hand between the two transducer from the position shown in the first draw-
transducers to absorb the sound, and you should see ing to the position shown in the second drawing, the
the amplitude of the signal from the receiver become student swept one trace past five complete cycles
much smaller. The receiving transducer will receive of the other trace. (The actual optical benches are
sound best at frequencies in the range of 35-40 kHz, about a meter long, not 8 or 9 cm as shown.) What
so keep the frequency in that range. is the wavelength of the ultrasound? [Self-check: you
should get 0.6 cm]
When you connect the function generator to both
the scope and the transmitting transducer, you’ll P2 Does it matter which transducer you move?
probably end up connecting a BNC-to-banana con-
P3 You can choose through how many wavelengths
nector to the function generator, and then putting a
you will move the transducer. What effect will this
second banana connector into the back of the first.
have on the accuracy of your determination of the
It’s important to make sure that the little tabs marked
speed of sound?
“GND” are on the same side of both connectors.
P4 What is a reasonable value for the speed of
If you are still having problems after taking the above
sound?
steps, try replacing one of the transducers — some
of the transducers are unreliable.
Self-Check
Observations Do an analysis without error bars before leaving lab,
and check that your speed of sound is reasonable.
Determine the wavelength and frequency of the sound
waves using the oscilloscope. Find out the tempera-
ture in the lab. Analysis
Do a quick analysis, without error analysis, during
Determine the speed of sound from your data, and
lab, to see if your result is reasonable.
use the techniques discussed in appendix 3 to derive
error bars.
Prelab Compare your result with the previously determined
value of √
The point of the prelab questions is to make sure
v = (20.1) T ,
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
where v is in m/s and T is the absolute temperature,
you don’t know the answers, make sure to come to
found by adding 273 to the Celsius temperature. Is
my office hours before lab and get help! Otherwise
it statistically consistent with your value?
you’re just setting yourself up for failure in lab.
P1 The drawings show two different configurations
of the transducers on the optical bench.

The scales are in cm. By sliding the right-hand

77
26 Electrical Resistance
Apparatus
DC power supply (Thornton) . . . . . . . . . . . . . 1/group
digital multimeters (Fluke and HP) . . . . . . . 2/group
resistors, various values
unknown electrical components
electrode paste
alligator clips
spare fuses for multimeters — Let students replace
fuses themselves.
Goals
Measure curves of voltage versus current for
three objects: your body and two unknown
electrical components.
Determine whether they are ohmic, and if so,
determine their resistances.
Introduction
Your nervous system depends on electrical currents,
and every day you use many devices based on elec-
trical currents without even thinking about it. De-
spite its ordinariness, the phenomenon of electric
currents passing through liquids (e.g., cellular flu-
ids) and solids (e.g., copper wires) is a subtle one.
For example, we now know that atoms are composed
of smaller, subatomic particles called electrons and
nuclei, and that the electrons and nuclei are elec-
trically charged, i.e., matter is electrical. Thus, we
now have a picture of these electrically charged par-
ticles sitting around in matter, ready to create an
electric current by moving in response to an exter-
nally applied voltage. Electricity had been used for
practical purposes for a hundred years, however, be-
fore the electrical nature of matter was proven at the
turn of the 20th century.
Another subtle issue involves Ohm’s law,
∆V
I= , less the length is so great that the resistance of the
R wire becomes non-negligible), and the schematic is
where ∆V is the voltage difference applied across an
object (e.g., a wire), and I is the current that flows
in response. A piece of copper wire, for instance,
has a constant value of R over a wide range of volt-
ages. Such materials are called ohmic. Materials
78 Lab 26 Electrical Resistance
with non-constant are called non-ohmic. The inter-
esting question is why so many materials are ohmic.
Since we know that electrons and nuclei are bound
together to form atoms, it would be more reasonable
to expect that small voltages, creating small electric
fields, would be unable to break the electrons and
nuclei away from each other, and no current would
flow at all — only with fairly large voltages should
the atoms be split up, allowing current to flow. Thus
we would expect R to be infinite for small voltages,
and small for large voltages, which would not be
ohmic behavior. It is only within the last 50 years
that a good explanation has been achieved for the
strange observation that nearly all solids and liquids
are ohmic.
Terminology, Schematics, and Re-
sistor Color Codes
The word “resistor” usually implies a specific type
of electrical component, which is a piece of ohmic
material with its shape and composition chosen to
give a desired value of R. Any piece of an ohmic
substance, however, has a constant value of R, and
therefore in some sense constitutes a “resistor.” The
wires in a circuit have electrical resistance, but the
resistance is usually negligible (a small fraction of an
Ohm for several centimeters of wire).
The usual symbol for a resistor in an electrical schematic
is this , but some recent schematics use
this . The symbol represents a fixed
source of voltage such as a battery, while repre-
sents an adjustable voltage source, such as the power
supply you will use in this lab.
In a schematic, the lengths and shapes of the lines
representing wires are completely irrelevant, and are
usually unrelated to the physical lengths and shapes
of the wires. The physical behavior of the circuit
does not depend on the lengths of the wires (un-
not meant to give any information other than that
needed to understand the circuit’s behavior. All that
really matters is what is connected to what.
For instance, the schematics (a) and (b) above are
completely equivalent, but (c) is different. In the
first two circuits, current heading out from the bat-
tery can “choose” which resistor to enter. Later on,
the two currents join back up. Such an arrangement
is called a parallel circuit. In the bottom circuit, a
series circuit, the current has no “choice” — it must
first flow through one resistor and then the other.
Resistors are usually too small to make it convenient
to print numerical resistance values on them, so they
are labeled with a color code, as shown in the table
and example below.
Setup
Obtain your two unknowns from your instructor.
Group 1 will use unknowns 1A and 1B, group 2 will
use 2A and 2B, and so on.
Here is a simplified version of the basic circuit you
will use for your measurements of I as a function of
∆V . Although I’ve used the symbol for a resistor,
the objects you are using are not necessarily resis-
tors, or even ohmic.
Here is the actual circuit, with the meters included.
In addition to the unknown resistance RU , a known
resistor RK (∼ 1kΩ is fine) is included to limit the
possible current that will flow and keep from blow-
ing fuses or burning out the unknown resistance with
too much current. This type of current-limiting ap-
plication is one of the main uses of resistors.
Observations
A Unknown component A
Set up the circuit shown above with unknown com-
ponent A. Most of your equipment accepts the ba-
nana plugs that your cables have on each end, but
to connect to RU and RK you need to stick alligator
clips on the banana plugs. See Appendix 7 for in-
formation about how to set up and use the two mul-
timeters. Do not use the pointy probes that come
with the multimeters, because there is no convenient
way to attach them to the circuit — just use the ba-
nana plug cables. Note when you need three wires to
79
come together at one point, you can plug a banana Now interpret the following color code:
plug into the back of another banana plug.
green orange yellow silver =?
Measure I as a function of ∆V . Make sure to take
P2 Fit a line to the following sample data and use
measurements for both positive and negative volt-
the slope to extract the resistance (see Appendix 4).
ages.
Often when we do this lab, it’s the first time in sev-
eral months that the meters have been used. The
small hand-held meters have a battery, which may
be dead. Check the battery icon on the LCD screen.

B Unknown component B
Repeat for unknown component B.

C The human body

Now do the same with the body of one member of
your group. This is not dangerous — the maxi-
mum voltage available from your power supply is
not enough to hurt you. (Children usually figure
out at some point that touching the terminals of a 9
V battery to their tongue gives an interesting sensa-
tion. The currents you will use in this lab are ten to Your result should be consistent with a resistor color
a hundred times smaller.) You may wish to keep the code of green-violet-yellow.
voltage below about 5 V or so. At voltages much P3 Plan how you will measure I versus ∆V for
higher than that (10 to 12 V), a few subjects get both positive and negative values of ∆V , since the
irritated skin. power supply only supplies positive voltages.
You will not want to use the alligator clips. With P4 Would data like these indicate a negative resis-
the power supply turned off, put small dabs of the tance, or did the experimenter just hook something
electrode paste on the subject’s left wrist and just up wrong? If the latter, explain how to fix it.
below the elbow, and simply lay the banana plug
connectors in the paste. The subject should avoid
moving. The paste is necessary because without it,
most of the resistance would come from the connec-
tion through the dry epidermal skin layer, and the
resistance would change erratically. The paste is a
relatively good conductor, and makes a better elec-
trical connection.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If P5 Explain why the following statements about
you don’t know the answers, make sure to come to the resistor RK are incorrect:
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab. a) “You have to make RK small compared to RU , so
it won’t affect things too much.”
P1 Check that you understand the interpretations
of the following color-coded resistor labels: b) “RK doesn’t affect the measurement of RU , be-
cause the meters just measure the total amount the
blue gray orange silver = 68 kΩ ± 10% power supply is putting out.”
blue gray orange gold = 68 kΩ ± 5%
blue gray red silver = 6.8 kΩ ± 10% c) “RK doesn’t affect the measurement of RU , be-
black brown blue silver = 1 MΩ ± 10% cause the current and voltage only go through RK
after they’ve already gone through RU .”

80 Lab 26 Electrical Resistance

Analysis
Graph I versus ∆V for all three unknowns. Decide
which ones are ohmic and which are non-ohmic. For
the ones that are ohmic, extract a value for the resis-
tance (see appendix 4). Don’t bother with analysis
of random errors, because the main source of error in
this lab is the systematic error in the calibration of
the multimeters (and in part C the systematic error
from the subject’s fidgeting).

Programmed Introduction to Prac-

tical Electrical Circuits
Physics courses in general are compromises between
the fundamental and the practical, between explor-
ing the basic principles of the physical universe and
developing certain useful technical skills. Although
the electricity and magnetism labs in this manual
are structured around the sequence of abstract the-
oretical concepts that make up the backbone of the
lecture course, it’s important that you develop cer-
tain practical skills as you go along. Not only will
they come in handy in real life, but the later parts
of this lab manual are written with the assumption
that you will have developed them.
As you progress in the lab course, you will find that
the instructions on how to construct and use circuits
become less and less explicit. The goal is not to
make you into an electronics technician, but neither
should you emerge from this course able only to flip
the switches and push the buttons on prepackaged
consumer electronics. To use a mechanical analogy,
the level of electrical sophistication you’re intended
to reach is not like the ability to rebuild a car engine
but more like being able to check your own oil.
In addition to the physics-based goals stated at the
beginning of this section, you should also be devel-
oping the following skills in lab this week:
(1) Be able to translate back and forth between schemat-
ics and actual circuits.
(2) Use a multimeter (discussed in Appendix 7),
given an explicit schematic showing how to connect
it to a circuit.
Further practical skills will be developed in the fol-
lowing lab.

81
27 The Loop and Junction Rules
Apparatus
DC power supply (Thornton) . . . . . . . . . . . . . 1/group
multimeter (Fluke) . . . . . . . . . . . . . . . . . . . . . . . 1/group
resistors
Goal
Test the loop and junction rules in two electrical
circuits.
Introduction
If you ask physicists what are the most fundamen-
tally important principles of their science, almost all
of them will start talking to you about conserva-
tion laws. A conservation law is a statement that a
certain measurable quantity cannot be changed. A
conservation law that is easy to understand is the
conservation of mass. No matter what you do, you
cannot create or destroy mass.
The two conservation laws with which we will be
concerned in this lab are conservation of energy and
conservation of charge. Energy is related to voltage,
because voltage is defined as V = P E/q. Charge
is related to current, because current is defined as
I = ∆q/∆t.
Conservation of charge has an important consequence
for electrical circuits:
When two or more wires come together at a point in
a DC circuit, the total current entering that point
equals the total current leaving it.
Such a coming-together of wires in a circuit is called
a junction. If the current leaving a junction was,
say, greater than the current entering, then the junc-
tion would have to be creating electric charge out
of nowhere. (Of course, charge could have been
stored up at that point and released later, but then
it wouldn’t be a DC circuit — the flow of current
would change over time as the stored charge was
used up.)
Conservation of energy can also be applied to an
electrical circuit. The charge carriers are typically
electrons in copper wires, and an electron has a po-
tential energy equal to −eV . Suppose the electron
sets off on a journey through a circuit made of re-
82 Lab 27 The Loop and Junction Rules
sistors. Passing through the first resistor, our sub-
atomic protagonist passes through a voltage differ-
ence of ∆V1 , so its potential energy changes by −e∆V1 .
To use a human analogy, this would be like going up
a hill of a certain height and gaining some gravi-
tational potential energy. Continuing on, it passes
through more voltage differences, −e∆V2 , −e∆V3 ,
and so on. Finally, in a moment of religious tran-
scendence, the electron realizes that life is one big
circuit — you always end up coming back where you
started from. If it passed through N resistors be-
fore getting back to its starting point, then the total
change in its potential energy was
−e (∆V1 + . . . + ∆VN ) .
But just as there is no such thing as a round-trip
hike that is all downhill, it is not possible for the
electron to have any net change in potential energy
after passing through this loop — if so, we would
have created some energy out of nothing. Since the
total change in the electron’s potential energy must
be zero, it must be true that ∆V1 + . . . + ∆VN = 0.
This is the loop rule:
The sum of the voltage differences around any closed
loop in a circuit must equal zero.
When you are hiking, there is an important distinc-
tion between uphill and downhill, which depends en-
tirely on which direction you happen to be traveling
on the trail. Similarly, it is important when apply-
ing the loop rule to be consistent about the signs
you give to the voltage differences, say positive if
the electron sees an increase in voltage and negative
if it sees a decrease along its direction of motion.
Observations
A The junction rule
Construct a circuit like the one in the figure, using
the Thornton power supply as your voltage source.
To make things more interesting, don’t use equal
resistors. Use nice big resistors (say 100 kΩ to 1
MΩ) — this will ensure that you don’t burn up the
resistors, and that the multimeter’s small internal
resistance when used as an ammeter is negligible in
comparison. Insert your multimeter in the circuit to
measure all three currents that you need in order to
test the junction rule.
 Analysis
Discuss whether you think your observations agree
with the loop and junction rules, taking into account
systematic and random errors. If this is your first
time doing error analysis, read appendices 2 and 3.

Programmed Introduction to Prac-

B The loop rule
Now come up with a circuit to test the loop rule.
tical Electrical Circuits
Since the loop rule is always supposed to be true, it’s The following practical skills are developed in this
hard to go wrong here! Make sure that (1) you have lab:
at least three resistors in a loop, (2) the whole cir-
(1) Use a multimeter without being given an explicit
cuit is not just a single loop, and (3) you hook in the
schematic showing how to connect it to your circuit.
power supply in a way that creates non-zero voltage
This means connecting it in parallel in order to mea-
differences across all the resistors. Measure the volt-
sure voltages and in series in order to measure cur-
age differences you need to measure to test the loop
rents.
rule. Here it is best to use fairly small resistances, so
that the multimeter’s large internal resistance when (2) Use your understanding of the loop and junc-
used in parallel as a voltmeter will not significantly tion rules to simplify electrical measurements. These
reduce the resistance of the circuit. Do not use re- rules often guarantee that you can get the same cur-
sistances of less than about 100 Ω, however, or you rent or voltage reading by measuring in more than
may blow a fuse or burn up a resistor. one place in a circuit. In real life, it is often much
easier to connect a meter to one place than another,
and you can therefore save yourself a lot of trouble
Prelab using the rules rules.
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Draw a schematic showing where you will in-
sert the multimeter in the circuit to measure the
currents in part A.
P2 Invent a circuit for part B, and draw a schematic.
You need not indicate actual resistor values, since
you will have to choose from among the values actu-
ally available in lab.
P3 Pick a loop from your circuit, and draw a schematic
showing how you will attach the multimeter in the
circuit to measure the voltage differences in part B.
P4 Explain why the following statement is incor-
rect: “We found that the loop rule was not quite
true, but the small error could have been because
the resistor’s value was off by a few percent com-
pared to the color-code value.”

Self-Check
Do the analysis in lab.

83
28 Electric Fields
Apparatus
board and U-shaped probe ruler
DC power supply (Thornton)
multimeter
scissors
stencils for drawing electrode shapes on paper
Goals
To be better able to visualize electric fields and
understand their meaning.
To examine the electric fields around certain
charge distributions.
Introduction
By definition, the electric field, E, at a particular
point equals the force on a test charge at that point
divided by the amount of charge, E = F/q. We can
plot the electric field around any charge distribution
by placing a test charge at different locations and
making note of the direction and magnitude of the
force on it. The direction of the electric field at
any point P is the same as the direction of the force
on a positive test charge at P. The result would be
a page covered with arrows of various lengths and
directions, known as a “sea of arrows” diagram..
In practice, Radio Shack does not sell equipment for
preparing a known test charge and measuring the
force on it, so there is no easy way to measure elec-
tric fields. What really is practical to measure at any
given point is the voltage, V , defined as the elec-
trical energy (potential energy) that a test charge
would have at that point, divided by the amount
of charge (E/Q). This quantity would have units
of J/C (Joules per Coulomb), but for convenience
we normally abbreviate this combination of units as
volts. Just as many mechanical phenomena can be
described using either the language of force or the
language of energy, it may be equally useful to de-
scribe electrical phenomena either by their electric
fields or by the voltages involved.
Since it is only ever the difference in potential en-
ergy (interaction energy) between two points that
can be defined unambiguously, the same is true for
84 Lab 28 Electric Fields
voltages. Every voltmeter has two probes, and the
meter tells you the difference in voltage between the
two places at which you connect them. Two points
have a nonzero voltage difference between them if
it takes work (either positive or negative) to move
a charge from one place to another. If there is a
voltage difference between two points in a conduct-
ing substance, charges will move between them just
like water will flow if there is a difference in levels.
The charge will always flow in the direction of lower
potential energy (just like water flows downhill).
All of this can be visualized most easily in terms
of maps of constant-voltage curves (also known as
equipotentials); you may be familiar with topograph-
ical maps, which are very similar. On a topograph-
ical map, curves are drawn to connect points hav-
ing the same height above sea level. For instance, a
cone-shaped volcano would be represented by con-
centric circles. The outermost circle might connect
all the points at an altitude of 500 m, and inside it
you might have concentric circles showing higher lev-
els such as 600, 700, 800, and 900 m. Now imagine
a similar representation of the voltage surrounding
an isolated point charge. There is no “sea level”
here, so we might just imagine connecting one probe
of the voltmeter to a point within the region to
be mapped, and the other probe to a fixed refer-
ence point very far away. The outermost circle on
your map might connect all the points having a volt-
age of 0.3 V relative to the distant reference point,
and within that would lie a 0.4-V circle, a 0.5-V
circle, and so on. These curves are referred to as
constant-voltage curves, because they connect points
of equal voltage. In this lab, you are going to map
out constant-voltage curves, but not just for an iso-
lated point charge, which is just a simple example
like the idealized example of a conical volcano.
You could move a charge along a constant-voltage
curve in either direction without doing any work,
because you are not moving it to a place of higher
potential energy. If you do not do any work when
moving along a constant-voltage curve, there must
not be a component of electric force along the surface
(or you would be doing work). A metal wire is a
constant-voltage curve. We know that electrons in a
metal are free to move. If there were a force along
the wire, electrons would move because of it. In fact
the electrons would move until they were distributed
in such a way that there is no longer any force on
them. At that point they would all stay put and
then there would be no force along the wire and it
would be a constant-voltage curve. (More generally,
any flat piece of conductor or any three-dimensional
volume consisting of conducting material will be a
constant-voltage region.)
There are geometrical and numerical relationships
between the electric field and the voltage, so even
though the voltage is what you’ll measure directly
in this lab, you can also relate your data to electric
fields. Since there is not any component of elec-
tric force parallel to a constant-voltage curve, elec-
tric field lines always pass through constant-voltage
curves at right angles. (Analogously, a stream flow-
ing straight downhill will cross the lines on a topo-
graphical map at right angles.) Also, if you divide
the work equation (∆energy) = F d by q, you get
(∆energy)/q = (F/q)d, which translates into ∆V =
−Ed. (The minus sign is because V goes down when
some other form of energy is released.) This means
that you can find the electric field strength at a point
P by dividing the voltage difference between the two
constant-voltage curves on either side of P by the
distance between them. You can see that units of
V/m can be used for the E field as an alternative to
the units of N/C suggested by its definition — the
units are completely equivalent.
A simplified schematic of the apparatus, being used with
pattern 1 on page 86.
A photo of the apparatus, being used with pattern 3 on
page 86.
Method
The first figure shows a simplified schematic of the
apparatus. The power supply provides an 8 V volt-
age difference between the two metal electrodes, drawn
in black. A voltmeter measures the voltage differ-
ence between an arbitrary reference voltage and a
point of interest in the gray area around the elec-
trodes. The result will be somewhere between 0 and
8 V. A voltmeter won’t actually work if it’s not part
of a complete circuit, but the gray area is intention-
ally made from a material that isn’t a very good
insulator, so enough current flows to allow the volt-
meter to operate.
The photo shows the actual apparatus. The elec-
trodes are painted with silver paint on a detachable
board, which goes underneath the big board. What
you actually see on top is just a piece of paper on
which you’ll trace the equipotentials with a pen. The
voltmeter is connected to a U-shaped probe with a
metal contact that slides underneath the board, and
a hole in the top piece for your pen.
Turn your large board upside down. Find the small
detachable board with the parallel-plate capacitor
pattern (pattern 1 on page 86) on it, and screw it to
the underside of the equipotential board, with the
silver-painted side facing down toward the tabletop.
Use the washers to protect the silver paint so that it
doesn’t get scraped off when you tighten the screws.
Now connect the voltage source (using the provided
wires) to the two large screws on either side of the
board. Referring to Appendix 7 on how to use a
85
multimeter, connect the multimeter so that you can
measure the voltage difference across the terminals
of the voltage source. Adjust the voltage source to
give 8 volts.
If you press down on the board, you can slip the pa-
per between the board and the four buttons you see
at the corners of the board. Tape the paper to your
board, because the buttons aren’t very dependable.
There are plastic stencils in some of the envelopes,
and you can use these to draw the electrodes accu-
rately onto your paper so you know where they are.
The photo, for example, shows pattern 3 traced onto
the paper.
Now put the U-probe in place so that the top is
above the equipotential board and the bottom of it
is below the board. You will first be looking for
places on the pattern board where the voltage is one
volt — look for places where the meter reads 1.0 and
mark them through the hole on the top of your U-
probe with a pencil or pen. You should find a whole
bunch of places there the voltage equals one volt,
so that you can draw a nice constant-voltage curve
connecting them. (If the line goes very far or curves
strangely, you may have to do more.) You can then
repeat the procedure for 2 V, 3 V, and so on. Label
each constant-voltage curve. Once you’ve finished
tracing the equipotentials, everyone in your group
will need one copy of each of the two patterns you
do, so you will need to photocopy them or simply
trace them by hand.
Repeat this procedure with another pattern. Groups
1 and 4 should do patterns 1 and 2; groups 2 and 5
patterns 1 and 3; groups 3, 6, and 7 patterns 1 and
4.
86 Lab 28 Electric Fields
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Looking at a plot of constant-voltage curves,
how could you tell where the strongest electric fields
would be? (Don’t just say that the field is strongest
when you’re close to “the charge,” because you may
have a complex charge distribution, and we don’t
have any way to see or measure the charge distribu-
tion.)
P2 What would the constant-voltage curves look
like in a region of uniform electric field (i.e., one in
which the E vectors are all the same strength, and
all in the same direction)?
Self-Check
Calculate at least one numerical electric field value
to make sure you understand how to do it.
You have probably found some constant-voltage curves
that form closed loops. Do the electric field patterns
ever seem to close back on themselves? Make sure
you understand why or why not.
Make sure the people in your group all have a copy
of each pattern.
Analysis
A. After you have completed the plots for two pat-
terns, you should try to draw in electric field vectors.
You will then have two different representations of
the field superimposed on one another. Remember
that electric field vectors are always perpendicular
to constant-voltage curves. The electric field lines
point from high voltage to low voltage, just as the
force on a rolling ball points downhill.
B. Select at least five places on each plot and deter-
mine the electric field strength (E) at each of them.
Make sure to include the two points that appear to
have the strongest and weakest fields.
C. For the parallel-plate capacitor, in what region
was the electric field relatively uniform?
87
29 Magnetism (Physics 206/211)
Apparatus
bar magnet (stack of 6 Nd)
compass
graph paper, with 1 cm squares
Hall effect magnetic field probes
LabPro interfaces, DC power supplies, and USB ca-
bles
Goal
Find how the magnetic field of a bar magnet changes
with distance along one of the magnet’s lines of sym-
metry.
Introduction
A Qualitative Mapping of the Magnet’s Field
You can use a compass to map out part of the mag-
netic field of a bar magnet. The compass is affected
by both the earth’s field and the bar magnet’s field,
and points in the direction of their vector sum, but if
you put the compass within a few cm of the bar mag-
net, you’re seeing mostly its field, not the earth’s.
Investigate the bar magnet’s field, and sketch it in
your lab notebook.
B Variation of Field With Distance: Deflection
of a Magnetic Compass
You can infer the strength of the bar magnet’s field
at a given point by putting the compass there and
seeing how much it is deflected.
The task can be simplified quite a bit if you restrict
yourself to measuring the magnetic field at points
along one of the magnet’s two lines of symmetry,
shown in the figure.
If the magnet is flipped across the vertical axis, the
north and south poles remain just where they were,
and the field is unchanged. That means the entire
magnetic field is also unchanged, and the field at a
point such as point b, along the line of symmetry,
must therefore point straight up.
If the magnet is flipped across the horizontal axis,
then the north and south poles are swapped, and the
field everywhere has to reverse its direction. Thus,
the field at points along this axis, e.g., point a, must
88 Lab 29 Magnetism (Physics 206/211)
point straight up or down.
Line up your magnet so it is pointing east-west.
Choose one of the two symmetry axes of your mag-
net, and measure the deflection of the compass at
two points along that axis, as shown in the second
figure, at the end of the lab. As part of your prelab,
you will use vector addition to find an equation for
Bm /Be , the magnet’s field in units of the Earth’s, in
terms of the deflection angle θ. For your first point,
find the distance r at which the deflection is 70 de-
grees; this angle is chosen because it’s about as big
as it can be without giving very poor relative preci-
sion in the determination of the magnetic field. For
your second data-point, use twice that distance. By
what factor does the field decrease when you double
r?
Note that the measurements are very sensitive to the
relative position and orientation of the bar magnet
and compass. You can position them accurately by
laying them both on top of a piece of graph paper,
but before you set all that up, get a preliminary
estimate of the distances you’ll be using, because
otherwise you can end up wasting your time.
Based on your two data-points, form a hypothesis
about the variation of the magnet’s field with dis-
tance according to a power law B ∝ rp .
C Variation of Field With Distance: Hall Effect
Magnetometer
In this part of the lab, you will test your hypothesis
about the power law relationship B ∝ rp ; you will
find out whether the field really does obey such a
law, and if it does, you will determine p accurately.
This part of the lab uses a device called a Hall ef- P2 Find Bm /Be in terms of the deflection angle θ.
fect magnetometer for measuring magnetic fields. It As a special case, you should be able to recover your
works by sending an electric current through a sub- answer to P1.
stance, and measuring the force exerted on those
moving charges by the surrounding magnetic field.
The probe only measures the component of the mag- Analysis
netic field vector that is parallel to its own axis. Plug
Determine the magnetic field of the bar magnet as
the probe into the LabPro interface, connect the in-
a function of distance. No error analysis is required.
terface to the computer’s USB port, and plug the in-
Look for a power-law relationship using the log-log
terface’s DC power supply in to it. Start up version
graphing technique described in appendix 5. Does
3 of Logger Pro, and it will automatically recognize
the power law hold for all the distances you investi-
the probe and start displaying magnetic fields on the
gated, or only at large distances?
screen, in units of mT (millitesla). The probe has
two ranges, one that can read fields up to 0.3 mT,
and one that goes up to 6.4 mT. You can select ei-
ther one using the switch on the probe. To test your
hypothesis with good precision, you need to obtain
data over the widest possible range of fields. Always
use the more sensitive 0.3 mT scale whenever possi-
ble, because it will give better precision for low fields.
Be careful, however, because if you expose the probe
to a field that’s beyond its maximum range, it will
give incorrect readings. Although you have an ex-
pectation about the direction of the field (based both
on symmetry arguments and on your qualitative re-
sults from part A), it’s a good idea to try orienting
the probe in different ways to see what happens.
Two extra complications are that the Earth’s field
is adding on to the magnet’s field, and the abso-
lute calibration of the probe is very poor by de-
fault. You can make the computer take care of both
of these issues automatically, by zeroing the sensor
(Experiment>Zero) when it is exposed only to the
Earth’s field. This causes the computer to impose a
calibration such that the Earth’s field is considered
to be exactly zero. You may need to redo the calibra-
tion each time you switch scales. If you then carry
out the whole measurement with the probe and the
magnet’s field both aligned east-west, the Earth’s
field has no effect.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 In part B, suppose that when the compass is
11.0 cm from the magnet, it is 45 degrees away from
north. What is the strength of the bar magnet’s field
at this location in space, in units of the Earth’s field?

89
Part B, measuring the variation of the bar magnet’s field with respect to distance

90 Lab 29 Magnetism (Physics 206/211)

91
30 The Dipole Field (Physics 222)
Apparatus
bar magnet
compass
graph paper, with 1 cm squares
Hall effect magnetic field probes
LabPro interfaces, DC power supplies, and USB ca-
bles
Goal
Find how the magnetic field of a bar magnet changes
with distance along one of the magnet’s lines of sym-
metry.
Introduction
This lab is designed to be used along with section
10.3 of Simple Nature, which is about the superpo-
sition (i.e., addition) of fields. That section is about
electric fields, and the basic principle is that if we
have two sets of sources (charges) that would indi-
vidually create fields E1 and E2 , then their combined
field is the vector sum E1 + E2 . Static electric fields,
however, are difficult to control and measure. Mag-
netic fields are much easier to work with, and the
same vector addition principle applies to them. In
this lab, you’ll expose a magnetic compass to the
superposed magnetic fields of the earth and a bar
magnet.
A Qualitative Mapping of the Dipole’s Field
You can use a compass to map out part of the mag-
netic field of a bar magnet. It turns out that the
bar magnet is the magnetic equivalent of an electric
dipole. The compass is affected by both the earth’s
field and the bar magnet’s field, and points in the
direction of their vector sum, but if you put the com-
pass within a few cm of the bar magnet, you’re seeing
mostly its field, not the earth’s. Investigate the bar
magnet’s field, and sketch it in your lab notebook.
You should see that it looks like the field a dipole.
B Variation of Field With Distance: Deflection
of a Magnetic Compass
Magnetic fields are actually measured in units of
Tesla (T), but for the purposes of this part of the lab,
we’ll just measure the fields in units of the earth’s
92 Lab 30 The Dipole Field (Physics 222)
magnetic field. That is, we define the earth’s mag-
netic field to have a strength of exactly 1.0 in Fuller-
ton.1 You can infer the strength of the bar magnet’s
field at a given point by putting the compass there
and seeing how much it is deflected. The standard
notation for magnetic field is B, so we can notate the
fields of the earth and the magnet as Be and Bm .
The task can be simplified quite a bit if you restrict
yourself to measuring the magnetic field at points
along one of the magnet’s two lines of symmetry,
shown in the figure.
If the magnet is flipped across the vertical axis, the
north and south poles remain just where they were,
and the field is unchanged. That means the entire
magnetic field is also unchanged, and the field at a
point such as point b, along the line of symmetry,
must therefore point straight up.
If the magnet is flipped across the horizontal axis,
then the north and south poles are swapped, and the
field everywhere has to reverse its direction. Thus,
the field at points along this axis, e.g., point a, must
point straight up or down.
Line up your magnet so it is pointing east-west.
Choose one of the two symmetry axes of your mag-
net, and measure the deflection of the compass at
two points along that axis, as shown in the second
figure, at the end of the lab. As part of your prelab,
you will use vector addition to find an equation for
Bm /Be , the magnet’s field in units of the Earth’s, in
terms of the deflection angle θ. For your first point,
1 Actually we’re defining its horizontal component to be
one unit — the compass can’t respond to vertical fields. The
dip angle of the magnetic field in Fullerton is fairly steep.
find the distance r at which the deflection is 70 de-
grees; this angle is choses because it’s about as big as
it can be without giving very poor relative precision
in the determination of the magnetic field. For your
second data-point, use twice that distance. By what
factor does the field decrease when you double r?
Note that the measurements are very sensitive to the
relative position and orientation of the bar magnet
and compass. You can position them accurately by
laying them both on top of a piece of graph paper,
but before you set all that up, get a preliminary
estimate of the distances you’ll be using, because
otherwise you can end up wasting your time.
Based on your two data-points, form a hypothesis
about the variation of the dipole’s field with dis-
tance according to a power law B ∝ rp . (If you’ve
done homework problems 11 and 16 in chapter 10 of
Simple Nature, then you know what p should be for
an electric dipole, based on vector addition of the
electric fields of two charges.)
C Variation of Field With Distance: Hall Effect
Magnetometer
In this part of the lab, you will test your hypothesis
about the power law relationship B ∝ rp ; you will
find out whether the field really does obey such a
law, and if it does, you will determine p accurately.
This part of the lab uses a device called a Hall effect
magnetometer for measuring magnetic fields. You
don’t know enough about magnetism yet to under-
stand the theory behind the operation of the de-
vice, so you can just think of it as a mysterious little
probe, like a wand, that you can place at some point
in space and measure the magnetic field. The probe
only measures the component of the magnetic field
vector that is parallel to its own axis. Plug the probe
into the LabPro interface, connect the interface to
the computer’s USB port, and plug the interface’s
DC power supply in to it. Start up version 3 of
Logger Pro, and it will automatically recognize the
probe and start displaying magnetic fields on the
screen, in units of mT (millitesla). The probe has
two ranges, one that can read fields up to 0.3 mT,
and one that goes up to 6.4 mT. You can select ei-
ther one using the switch on the probe. To test your
hypothesis with good precision, you need to obtain
data over the widest possible range of fields. Al-
ways use the more sensitive 0.3 mT scale whenever
possible, because it will give better precision for low
fields. Be careful, however, because if you expose the
probe to a field that’s beyond its maximum range, it
will give incorrect readings. Although you have an
expectation about the direction of the field (based
both on symmetry arguments and on your qualita-
tive results from part A), it’s a good idea to try
orienting the probe along different axes to see what
happens. In general, if you want to use the probe to
measure a field whose direction and magnitude are
both unknown, you need to orient the probe along
two different axes, and determine the two compo-
nents separately.
Two extra complications are that the Earth’s field
is adding on to the magnet’s field, and the abso-
lute calibration of the probe is very poor by de-
fault. You can make the computer take care of both
of these issues automatically, by zeroing the sensor
(Experiment>Zero) when it is exposed only to the
Earth’s field, and aligned perpendicular to it. This
causes the computer to impose a calibration such
that the Earth’s field is considered to be exactly zero.
You may need to redo the calibration each time you
switch scales. If you then carry out the whole mea-
surement with the probe and the magnet’s field both
aligned east-west, the Earth’s field has no effect.
D Variation of Field With Angle: Hall Effect Mag-
netometer
Homework problems 11 and 16 in chapter 10 of Sim-
ple Nature, predict that for an electric dipole, the
field in the midplane is exactly half as strong as the
on-axis field, at the same distance. Test this predic-
tion.
Also, find the magnitude of the field at an angle
of 45 degrees between the midplane and the axis.
Since you don’t know the direction of the field at
this location based on symmetry arguments (and you
only know it very roughly based on mapping with a
compass in part A), you’ll need to measure both of
the field’s components at this location.
As you plan your observations in this part, you’ll
need to think about what is the best distance at
which to place the probe. If the distance is too large,
you may find that the field is too weak to measure
with good precision. If the distance is too small, then
the physical size of the probe becomes an issue, since
the exact location at which the probe measures the
field is ill-defined. (The probe measures a voltage
created by the field in a sample of some material,
and that sample has a finite size.)
In part C, all the fields were along a single line, and
there were no angles involved. That made it simple
to get rid of the effect the Earth’s field. That doesn’t
work in this part, however. One way of handling the
difficulty is to flip the magnet by 180 degrees, and
find the difference between the readings for the two
93
opposite orientations of the magnet, which should
equal twice the magnet’s field. The Earth’s field can-
cels out. This means that you need a total of four
different measurements at each point in space, cover-
ing all four possible combinations of the orientation
of the probe along x or y with both orientations of
the magnet.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Suppose that when the compass is 11.0 cm from
the magnet, it is 45 degrees away from north. What
is the strength of the bar magnet’s field at this loca-
tion in space, in units of the Earth’s field?
P2 Find Bm /Be in terms of the deflection angle θ.
As a special case, you should be able to recover your
answer to P1.

Analysis
Determine the magnetic field of the bar magnet as
a function of distance. No error analysis is required.
Look for a power-law relationship using the log-log
graphing technique described in appendix 5. Does
the power law hold for all the distances you inves-
tigated, or only at large distances? Compare this
power law result with the result for the variation of
an electric dipole’s field with distance.

94 Lab 30 The Dipole Field (Physics 222)

Measuring the variation of the bar magnet’s field with respect to distance.

95
31 The Earth’s Magnetic Field (Physics 222)
Apparatus thread, and observe the period of its oscillations in
the Earth’s magnetic field. The idea is that if the
digital multimeter Earth’s field is stronger, there is a stronger torque
neodymium magnet (6 discs stuck together) trying to align the magnet north-south, and the fre-
magnetic compass quency of the oscillations will therefore be higher.
resistors By measuring the frequency of the oscillations, we
decade resistor boxes can work backward and infer the strength of the hor-
rulers izontal component of the Earth’s field.
thread
1-m aluminum rod
stopwatch
photogate
laser
aluminum rods, and clamps
D cell batteries and holders . . . . . . . . . . . . . . . 2/group
Helmholtz coils (e/m apparatus)
high-precision Helmholtz coil (one set)
Hall effect magnetic field probes
LabPro interfaces, DC power supplies, and USB ca-
bles

Goal
Determine the horizontal component of the Earth’s
magnetic field in Fullerton, to high precision.

Observations A contour map of the field of a Helmholtz coil (top view

Since you’ve already used the Hall effect magnetic
field probes in lab 30, you might think that it would
be relatively trivial to measure the Earth’s magnetic
field precisely. However, the calibration of those
probes is quite poor, so it’s not possible to get results
with error bars smaller than about 10-20%.
The geometry of a Helmholtz coil.
The basic idea of the more precise technique used
in this lab is to hang a permanent magnet from a
of the horizontal plane cutting through the center).
One reason the technique isn’t quite that simple is
that the frequency of the oscillations also depends
on other quantities, including the magnet’s dipole
moment and moment of inertia, that are very diffi-
cult to measure with better than about 10% preci-
sion. A trick for getting around this problem is to
superimpose a known southward magnetic field on
the Earth’s northward one, and adjust the known
field so as to cancel the Earth’s. Reducing the field
increases the period of the oscillations, and if we
could exactly cancel the horizontal component of the
Earth’s field, then the period would be infinite. The
known field is supplied by a type of electromagnet
called a Helmholtz coil, shown in the first figure. It
consists of two circular coils of wire, with their axes
coinciding. In the classic design (which is what’s
really properly called a Helmholtz coil), the separa-
tion h between the planes of the two coils is equal to
their radius, b. Having h = b turns out to produce

96 Lab 31 The Earth’s Magnetic Field (Physics 222)

the most uniform possible field near the center of the
whole arrangement, in the sense that all the field’s
derivatives up to the fourth derivative equal zero.
The second figure (from the Wikipedia article, copy-
left licensed by Wikipedia) is a contour map showing
how little the field actually varies over a fairly large
volume in the center. The “octopus” in the middle
is the region in which the field is between 99% and
101% of its value at the center.
Even this version of the experiment turns out to need
some further tweaking. It is difficult to align the axis
of the coils with the Earth’s field, so we typically end
up with a misalignment, φ, which is a few degrees.
Therefore, the fields do not really cancel, and as the
current through the coils is tuned through the op-
timal value, the horizontal field becomes small, but
not zero, and swings around gradually from north
to south. It becomes difficult to pick off the current
that produces the maximum period, partly because
the period of the oscillations is not quite indepen-
dent of amplitude, and it becomes difficult to con-
trol the amplitude of the oscillations properly when
the equilibrium orientation is constantly changing.
Even if we could precisely recognize the current that
gave the maximum period, that would be the current
that canceled out the component of the Earth’s field
along the coils’ axis, i.e., we would be taking the
vector (Bx , By ), and changing it to (Bx , 0), where
y is the axis of the coils. Thus we would really be
measuring, By = B cos φ, rather than |B|. To get
around this problem, you can use the following itera-
tive method: (1) Align the coils’ axis approximately
with the earth’s field by eyeballing the alignment
against a magnetic compass. (2) Tune the current
in the coil to the point where the magnet’s equilib-
rium orientation is perpendicular to the earth’s field.
This is pretty close to the current that would have
canceled the earth’s field, if the alignment had been
perfect. In this state, the magnet will point either to
the east or to the west, depending on the direction of
the error in alignment. (3) Carefully, slowly rotate
the apparatus until the magnet’s equilibrium orien-
tation shifts to the north-south line. This is a state
in which the coil’s field is exaclty on the same line
as the Earth’s, but their magnitudes are slightly dif-
ferent. (4) Tune the current again to maximize the
period. In this final step, it becomes important to
control the amplitude of the oscillations. As shown
in the figure, the error in the period is less than 0.1%
for amplitudes of less than about 10 degrees.
The problem now boils down to the accurate deter-
mination of the field at the center of the Helmholtz
coils for a given amount of current, i.e., the ratio
The dependence of the period on amplitude. For an-
gles less than 20 degrees, the motion is nearly simple
harmonic, and the period is independent of amplitude to
within about 1%. Higher amplitudes can be used, but it
becomes much more important to control the initial am-
plitude.
B/I. You’ll derive the relevant expression as one of
your prelab questions. It depends on the accurate
measurement of the dimensions b and h. In general
it’s fairly difficult to construct magnet coils so that
their dimensions are accurately determinable, and
the coils you’ll use are no exception. They consist
of somewhat irregular bundles of wire tied together
with cable ties, and they aren’t even circular; their
vertical diameter is significantly different from their
horizontal diameter. As closely as I’ve been able to
determine, they have h = 14.7 ± 0.3 cm, and an av-
erage b of about 15.1 ± 0.3 cm, but these error bars
are uncomfortably large. They have N = 130 turns
of wire on each coil, i.e., 260 turns on each complete
set of Helmholtz coils.
Because of these problems, I’ve constructed a Helm-
holtz coil that has a much more precisely measur-
able geometry. You can calculate B/I for the pre-
cise coils, whose dimensions are carefully constructed
and easy to measure: h = b = 11.15 ± .05 cm. They
have N = 5 turns of wire in each coil. Although
there is only one copy of the precise Helmholtz coil,
and it wouldn’t be convenient to use for this lab
anyway (they produce weak fields, and their interior
is not very accessible), we can calibrate your coils
against them. I’m planning to do this as a student
lab for the first time in spring 2009, and we’ll use
the data from that semester as a calibration for the
coils, by comparing Bearth /I for them with Bearth /I
for the precise coils.
The Helmholtz coils we’re using are actually meant
97
for lab 33, and they have a big, extremely expensive
vacuum tube stuck inside them for that purpose.
With your instructor’s help, very carefully detach
the base from the tube. Then unscrew the yokes
that hold the tube in place, and put the tube out of
the way in the stockroom, with cusioning to make
sure it doesn’t get broken.
You need precisely controlled, steady currents for
this lab, and DC power supplies aren’t stable enough,
so you’ll use batteries instead. To control the cur-
rent precisely, you’ll use the decade resistor boxes,
which are variable resistors that let you dial up any
decimal number of ohms that you want.
We want to keep all magnetic materials far away
from the magnet. Clamp the 1-m aluminum rod to
the vertical steel post, and hang the magnet from it,
far from the post.
There are several possible methods for measuring
the period of the oscillations, and one of my goals
for spring 2009 is to have my students test drive
them. One is to use a stopwatch to time, say, 20
oscillations. A second method would be to use the
magnetic field probe and graph the field as a function
of time. A third method would be to use a photo-
gate, in pendulum mode as described in appendix
6. The photogates have steel screws in them, so you
can’t use them in the ordinary way, with the mag-
net swinging through the infrared beam that goes
across the center of the gate. Instead, you can open
the shutter on the inside of the photogate to change
it into a mode where it senses light from the beam of
an external laser. The photogate can then be phys-
ically far away from the magnet so that the screws
don’t affect the measurement. A possible problem
with the photogate method is that it requires the
amplitude of the oscillations to be big enough so
that the magnet blocks and unblocks the photogate,
but with oscillations that big, the dependence of the
period on amplitude could be a significant source
of error unless the amplitude was very accurately
controlled. This problem could possibly be solved
by attaching a cardboard vane to the magnet, and
that would also get rid of the safety problem caused
by reflecting the laser beam from the shiny magnet.
Of these methods, it’s possible that one might be
the most convenient for rough initial measurements,
while another would work best for the final, accurate
measurement.
When you’re done with all this, what you’ve actually
measured is the magnetic field inside the building.
Many buildings have magnetic building materials, so
the fields inside them are different from the Earth’s
98 Lab 31 The Earth’s Magnetic Field (Physics 222)
field. To correct for this, measure the period of the
magnet’s oscillation inside and outside. If they are
significantly different, correct according to B1 /B2 =
2
(T2 /Tp1 ) ; this follows from adapting the equation
ω = k/m for simple harmonic motion to the case
of rotation, with the torque τ = m × B playing the
role of the restoring force.
(Note to myself: Some of my own further notes about
the lab are embedded in comments in the LaTeX
source code for the lab manual.)
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 For an electromagnet consisting of a single cir-
cular loop of wire of radius b, the field at a point on
its axis, at a distance z from the plane of the loop,
is given by
2πkIb2
B= .
c2 (b2 + z 2 )3/2
Starting from this equation, derive an equation for
the magnetic field at the center of a pair of Helmholtz
coils, in terms of h, b, and N . Find B/I for both
the high-precision coils and the low-precision ones,
based on the given values of h, b, and N . (The B/I
for the low-precision ones is useful as a check, but has
poor precision, which is why you’ll calibrate against
the high-precision ones.)
P2 Estimate the current that will be required in
the low-precision coils in order to cancel the Earth’s
field, about 2 × 10−5 T.
Analysis
Find the earth’s magnetic field, with error bars.
99
32 Relativity
Apparatus
magnetic balance . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
meter stick . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
multimeter (BK, not HP) . . . . . . . . . . . . . . . . . 1/group
laser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
vernier calipers . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
decade resistor box (General Radio) . . . . . . 1/group
staples
lab’s DC power supply (30 A)
Goal
Measure the speed of light.
Introduction
Oersted discovered that magnetism is an interac-
tion of moving charges with moving charges, but
it wasn’t until almost a hundred years later that
Einstein showed why such an interaction must exist:
magnetism occurs as a direct result of his theory of
relativity. Since magnetism is a purely relativistic
effect, and relativistic effects depend on the speed of
light, any measurement of a magnetic effect can be
used to determine the speed of light.
Setup
The idea is to set up opposite currents in two wires,
A and B, one under the other, and use the repulsion
between the currents to create an upward force on
the top wire, A. The top wire is on the arm of a bal-
ance, which has a stable equilibrium because of the
weight C hanging below it. You initially set up the
balance with no current through the wires, adjusting
the counterweight D so that the distance between the
wires is as small as possible. What we care about is
really the center-to-center distance (which we’ll call
R), so even if the wires are almost touching, there’s
still a millimeter or two worth of distance between
them.) By shining a laser at the mirror, E, and ob-
serving the spot it makes on the wall, you can very
accurately determine this particular position of the
balance, and tell later on when you’ve reproduced it.
If you put a current through the wires, it will raise
wire A. The torque made by the magnetic repulsion
100 Lab 32 Relativity
is now canceling the torque made by gravity directly
on all the hardware, such as the masses C and D.
This gravitational torque was zero before, but now
you don’t know what it is. The trick is to put a tiny
weight (a staple) on top of wire A, and adjust the
current so that the balance returns to the position it
originally had, as determined by the laser dot on the
wall. You now know that the gravitational torque
acting on the original apparatus (everything except
for the staple) is back to zero, so the only torques
acting are the torque of gravity on the staple and
the magnetic torque. Since both these torques are
applied at the same distance from the axis, the forces
creating these torques must be equal as well. By
weighing a block of staples, you can determine the
weight of one staple, and infer the magnetic force
that was acting.
It’s very important to get the wires A and B per-
fectly parallel. You also need to minimize the resis-
tance of the apparatus, or else you won’t be able to
get enough current through it to cancel the weight
of the staple. Most of the resistance is at the pol-
ished metal knife-edges that the moving part of the
balance rests on. It may be necessary to clean the
surfaces, or even to freshen them a little with a file
to remove any layer of oxidation. Since everyone
is sharing the same power supply, you can’t turn a
knob to control the voltage being applied to your
setup. Instead, you need to put the decade resistor
box in series in your circuit, and use it to control the
current that flows.
Analysis
The mass of an aluminum atom is 4.48 × 10−26 kg.
Let’s assume that each aluminum atom contributes
one conduction electron, and that the wires have law to a cylinder of radius R and length `:
masses per unit length of 2 g/m — these two as-
sumptions are only roughly right, but you’ll see later ΦE = 4πkqin
that they end up not mattering. (E)(2πR`) = 4πkλtotal `
You can now calculate the number of coulombs per 2kλtotal
E=
meter of conduction electrons, −λ, in your wires. By R
combining this with your measured levitation cur- The electrical force Eq = Eλ` cancels out the grav-
rent, you can find the average velocity, v, at which itational force mg acting on the staple, so ignoring
the electrons were drifting through the wire. This ve- plus and minus signs, we have
locity is quite small compared to the speed of light,
so the relativistic effect is slight. However, as you Eλ` = mg
found when you did the prelab, the amount of charge
6kλ2 v 2 mg
in a piece of ordinary matter is huge, so even a slight =
4Rc2 `
effect is enough to produce a measurable result.
Now imagine yourself as one of the moving electrons But λv is just the current, so
in the top wire. In your frame of reference, the elec-
trons in the other strip are moving at velocity −2v, 6kI 2 mg
=
and for each such electron there is a corresponding 4Rc2 `
proton moving at velocity −v relative to you. (You
don’t care about the protons and electrons that are
paired off in atoms, because they cancel each other.) Solving for c, we have
Both the electrons and the protons are squashed to- s
gether by the relativistic contraction of space, so we 6k`
have c=I
4Rgm
1
λp = λ p Note that although I asked you to calculate v and
1 − v 2 /c2 λ for physical insight, it turns out that all you re-
1 ally need to know is their product, which equals the
λe = −λ p .
1 − (2v)2 /c2 current you read on your meter.
Your final result is the speed of light, with error bars.
In the frame of reference fixed to the tabletop, these
would have canceled each other out, but in your
frame of reference, we have Prelab
λtotal = λp + λe The point of the prelab questions is to make sure
" # you understand what you’re doing, why you’re do-
1 1 ing it, and how to avoid some common mistakes. If
=λ p −p
1− v 2 /c2 1 − (2v)2 /c2 you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
You may want to try calculating this directly just you’re just setting yourself up for failure in lab.
for fun, but unless your calculator has unusually Do the laser safety checklist, Appendix 9, tear it out,
high precision, it will round off to zero, since the and turn it in at the beginning of lab. If you don’t
gamma factors are both very close to one. To get understand something, don’t initial that point, and
a useful result, we need to use the approximation ask your instructor for clarification before you start
(1 − )−1/2 ≈ 1 + /2, which results in the lab.
P1 Calculate −λ, the number of coulombs per me-
3v 2 ter in the tabletop’s frame of reference, using the
λtotal ≈ −λ 2 .
4c assumptions given above. Answer: −7 × 103 C/m
In your frame of reference, the electric field of this P2 This is a huge amount of charge! Why doesn’t
charge is what is responsible for repelling you and it produce any measurable electrical forces when the
causing the upward electric force on the wire. The wire is just lying there without being connected to
electric force can be calculated by applying Gauss’ any electrical circuit?

101
33 The Charge to Mass Ratio of the Electron
Apparatus
vacuum tube with Helmholtz
coils (Leybold ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Cenco 33034 HV supply . . . . . . . . . . . . . . . . . . . . . . . . . 1
12-V DC power supplies (Thornton) . . . . . . . . . . . . . 1
multimeters (Fluke or HP) . . . . . . . . . . . . . . . . . . . . . . 2
compass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ruler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
banana-plug cables

Goal
Measure the charge-to-mass ratio of the electron.
of the magnetic field on the electrons is

F = qvB , (1)
Introduction
Why should you believe electrons exist? By the turn directed towards the center of the circle. Their ac-
of the twentieth century, not all scientists believed celeration is
in the literal reality of atoms, and few could imag-
ine smaller objects from which the atoms themselves v2
were constructed. Over two thousand years had a= , (2)
r
elapsed since the Greeks first speculated that atoms
existed based on philosophical arguments without so using F = ma, we can write
experimental evidence. During the Middle Ages in
Europe, “atomism” had been considered highly sus- mv 2
qvB = . (3)
pect, and possibly heretical. Finally by the Vic- r
torian era, enough evidence had accumulated from
If the initial velocity of the electrons is provided by
chemical experiments to make a persuasive case for
accelerating them through a voltage difference V ,
atoms, but subatomic particles were not even dis-
they have a kinetic energy equal to qV , so
cussed.
If it had taken two millennia to settle the question 1
mv 2 = qV . (4)
of atoms, it is remarkable that another, subatomic 2
level of structure was brought to light over a period
From equations 3 and 4, you can determine q/m.
of only about five years, from 1895 to 1900. Most
Note that since the force of a magnetic field on a
of the crucial work was carried out in a series of
moving charged particle is always perpendicular to
experiments by J.J. Thomson, who is therefore often
the direction of the particle’s motion, the magnetic
considered the discoverer of the electron.
field can never do any work on it, and the particle’s
In this lab, you will carry out a variation on a crucial KE and speed are therefore constant.
experiment by Thomson, in which he measured the
You will be able to see where the electrons are going,
ratio of the charge of the electron to its mass, q/m.
because the vacuum tube is filled with a hydrogen
The basic idea is to observe a beam of electrons in
gas at a low pressure. Most electrons travel large
a region of space where there is an approximately
distances through the gas without ever colliding with
uniform magnetic field, B. The electrons are emitted
a hydrogen atom, but a few do collide, and the atoms
perpendicular to the field, and, it turns out, travel
then give off blue light, which you can see. Although
in a circle in a plane perpendicular to it. The force
I will loosely refer to “seeing the beam,” you are
really seeing the light from the collisions, not the

102 Lab 33 The Charge to Mass Ratio of the Electron

beam of electrons itself. The manufacturer of the
tube has put in just enough gas to make the beam
visible; more gas would make a brighter beam, but
would cause it to spread out and become too broad
to measure it precisely.
The field is supplied by an electromagnet consisting
of two circular coils, each with 130 turns of wire
(the same on all the tubes we have). The coils are
placed on the same axis, with the vacuum tube at
the center. A pair of coils arranged in this type of
geometry are called Helmholtz coils. Such a setup
provides a nearly uniform field in a large volume
of space between the coils, and that space is more
accessible than the inside of a solenoid.
Safety
You will use the Cenco high-voltage supply to make
a DC voltage of about 300 V . Two things automat-
ically keep this from being very dangerous:
Several hundred DC volts are far less danger-
ous than a similar AC voltage. The household
AC voltages of 110 and 220 V are more dan-
gerous because AC is more readily conducted
by body tissues.
The HV supply will blow a fuse if too much
current flows.
Do the high voltage safety checklist, Appendix 8,
tear it out, and turn it in at the beginning of lab. If
you don’t understand something, don’t initial that
point, and ask your instructor for clarification before
you start the lab.
Setup
Before beginning, make sure you do not have any
computer disks near the apparatus, because the mag-
netic field could erase them.
Heater circuit: As with all vacuum tubes, the cath-
ode is heated to make it release electrons more easily.
There is a separate low-voltage power supply built
into the high-voltage supply. It has a set of plugs
that, in different combinations, allow you to get var-
ious low voltage values. Use it to supply 6 V to the
terminals marked “heater” on the vacuum tube. The
tube should start to glow.
Electromagnet circuit: Connect the other Thornton
power supply, in series with an ammeter, to the ter-
minals marked “coil.” The current from this power
supply goes through both coils to make the magnetic
field. Verify that the magnet is working by using it
to deflect a nearby compass.
High-voltage circuit: Leave the Cenco HV supply
unplugged. It is really three HV circuits in one box.
You’ll be using the circuit that goes up to 500 V.
Connect it to the terminals marked “anode.” Ask
your instructor to check your circuit. Now plug in
the HV supply and turn up the voltage to 300 V .
You should see the electron beam. If you don’t see
anything, try it with the lights dimmed.
Observations
Make the necessary observations in order to find
q/m, carrying out your plan to deal with the effects
of the Earth’s field. The high voltage is supposed
to be 300 V, but to get an accurate measurement
of what it really is you’ll need to use a multimeter
rather than the poorly calibrated meter on the front
of the high voltage supply.
The beam can be measured accurately by using the
glass rod inside the tube, which has a centimeter
scale marked on it.
Be sure to compute q/m before you leave the lab.
That way you’ll know you didn’t forget to measure
something important, and that your result is reason-
able compared to the currently accepted value.
There is a glass rod inside the vacuum tube with a
centimeter scale on it, so you can measure the diam-
eter d of the beam circle simply by looking at the
place where the glowing beam hits the scale. This is
much more accurate than holding a ruler up to the
tube, because it eliminates the parallax error that
would be caused by viewing the beam and the ruler
along a line that wasn’t perpendicular to the plane of
the beam. However, the manufacturing process used
in making these tubes (they’re probably hand-blown
by a glass blower) isn’t very precise, and on many of
the tubes you can easily tell by comparison with the
a ruler that, e.g., the 10.0 cm point on the glass rod
is not really 10.0 cm away from the hole from which
the beam emerges. Past students have painstakingly
determined the appropriate corrections, k, to add to
the observed diameters by the following electrical
method. If you look at your answer to prelab ques-
tion P1, you’ll see that the product Br is always a
fixed quantity in this experiment. It therefore fol-
lows that Id is also supposed to be constant. They
measured I and d at two different values of I, and
determined the correction k that had to be added to
their d values in order to make the two values of Id
103
equal. The results are as follows: coils. Devise a plan to eliminate, correct for, or at
least estimate the effect of the Earth’s magnetic field
serial number k (cm)
on your final q/m value.
98-16 0.0
99-08 -0.6 P5 Of the three circuits involved in this experi-
99-10 -0.2 ment, which ones need to be hooked up with the
99-17 +0.2 right polarity, and for which ones is the polarity ir-
99-56 +0.3 relevant?
If your apparatus is one that hasn’t already had its k P6 What would you infer if you found the beam
determined, then you should do the necessary mea- of electrons formed a helix rather than a circle?
surements to calibrate it.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
The week before you are to do the lab, briefly famil-
iarize yourself visually with the apparatus.
Do the high voltage safety checklist, Appendix 8,
tear it out, and turn it in at the beginning of lab. If
you don’t understand something, don’t initial that
point, and ask your instructor for clarification before
you start the lab.
Analysis
Determine q/m, with error bars.
Answer the following questions:
Q1. Thomson started to become convinced during
his experiments that the “cathode rays” observed
coming from the cathodes of vacuum tubes were
building blocks of atoms — what we now call elec-
trons. He then carried out observations with cath-
odes made of a variety of metals, and found that
q/m was the same in every case. How would that
observation serve to test his hypothesis?
Q2. Why is it not possible to determine q and m
themselves, rather than just their ratio, by observing
electrons’ motion in electric or magnetic fields?

P1 Derive an equation for q/m in terms of V , r
and B.
P2 For an electromagnet consisting of a single cir-
cular loop of wire of radius b, the field at a point on
its axis, at a distance z from the plane of the loop,
is given by
Q3. Thomson found that the q/m of an electron
was thousands of times larger than that of ions in
electrolysis. Would this imply that the electrons had
more charge? Less mass? Would there be no way to
tell? Explain.

2πkIb2
B= .
c2 (b2+ z 2 )3/2
Starting from this equation, derive an equation for
the magnetic field at the center of a pair of Helmholtz
coils. Let the number of turns in each coil be N (in
our case, N = 130), let their radius be b, and let the
distance between them be h. (In the actual experi-
ment, the electrons are never exactly on the axis of
the Helmholtz coils. In practice, the equation you
will derive is sufficiently accurate as an approxima-
tion to the actual field experienced by the electrons.)
If you have trouble with this derivation, see your in-
structor in his/her office hours.
P3 Find the currently accepted value of q/m for
the electron.
P4 The electrons will be affected by the Earth’s
magnetic field, as well as the (larger) field of the

104 Lab 33 The Charge to Mass Ratio of the Electron

105
34 Energy in Fields
Apparatus
Heath coils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
0.01 µF capacitors . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Daedalon function generator . . . . . . . . . . . . . . 1/group
PASCO PI-9587C sine-wave generator . . . . 1/group
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Goal
Observe how the energy content of a field relates to
the field strength.
Introduction
A simplified version of the circuit.
The basic idea of this lab is to observe a circuit like
the one shown in the figure above, consisting of a ca-
pacitor and a coil of wire (inductor). Imagine that
we first deposit positive and negative charges on the
plates of the capacitor. If we imagined that the uni-
verse was purely mechanical, obeying Newton’s laws
of motion, we would expect that the attractive force
between these charges would cause them to come
back together and reestablish a stable equilibrium
in which there was zero net charge everywhere in
the circuit.
However, the capacitor in its initial, charged, state
has an electric field between its plates, and this field
possesses energy. This energy can’t just go away,
because energy is conserved. What really happens
is that as charge starts to flow off of the capacitor
plates, a current is established in the coil. This cur-
rent creates a magnetic field in the space inside and
around the coil. The electric energy doesn’t just
evaporate; it turns into magnetic energy. We end
up with an oscillation in which the capacitor and
the coil trade energy back and forth. Your goal is
to monitor this energy exchange, and to use it to
deduce a power-law relationship between each field
and its energy.
106 Lab 34 Energy in Fields
The actual circuit.
The practical realization of the circuit involves some
further complications, as shown in the second figure.
The wires are not superconductors, so the circuit has
some nonzero resistance, and the oscillations would
therefore gradually die out, as the electric and mag-
netic energies were converted to heat. The sine wave
generator serves both to initiate the oscillations and
to maintain them, replacing, in each cycle, the en-
ergy that was lost to heat.
Furthermore, the circuit has a resonant frequency
at it prefers to oscillate, and when the resistance is
very small, the width of the resonance is very nar-
row. To make the resonance wider and less finicky,
we intentionally insert a 10 kΩ resistor. The induc-
tance of the coil is about 1 H, which gives a resonant
frequency of about 1.5 kHz.
The actual circuit consists of the 1 H Heath coil, a
0.01 µF capacitance supplied by the decade capaci-
tor box, a 10 kΩ resistor, and the PASCO sine wave
generator (using the GND and LO Ω terminals).
Observations
Let E be the magnitude of the electric field between
the capacitor plates, and let Ẽ be the maximum
value of this quantity. It is then convenient to define
x = E/Ẽ, a unitless quantity ranging from −1 to 1.
Similarly, let y = B/B̃ for the corresponding mag-
netic quantities. The electric field is proportional
to the voltage difference across the capacitor plates,
which is something we can measure directly using
the oscilloscope:
E VC
x= =
Ẽ V˜C
Magnetic fields are created by moving charges, i.e.,
by currents. Unfortunately, an oscilloscope doesn’t
measure current, so there’s no equally direct way to
get a handle on the magnetic field. However, all
the current that goes through the coil must also go
through the resistor, and Ohm’s law relates the cur-
rent through the resistor to the voltage drop across
it. This voltage drop is something we can measure Prelab
with the oscilloscope, so we have
The point of the prelab questions is to make sure
B I VR you understand what you’re doing, why you’re do-
y= = =
B̃ ˜
I V˜R ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
To measure x and y, you need to connect channels my office hours before lab and get help! Otherwise
1 and 2 of the oscilloscope across the resistor and you’re just setting yourself up for failure in lab.
the capacitor. Since both channels of the scope are
grounded on one side (the side with the ground tab P1 Sketch what your graph would look like for
on the banana-to-bnc connector), you need to make p = 0.1, p = 1, p = 2, and p = 10. (You should
sure that their grounded sides both go to the piece of be able to do p = 1 and p = 2 without any compu-
wire between the resistor and the capacitor. Further- tations. For p = 0.1 and p = 10, you can either run
more, one output of the sine wave generator is nor- some numbers on your calculator or use your math-
mally grounded, which would mess everything up: ematical knowledge to sketch what they would turn
two different points in the circuit would be grounded, out like.)
which would mean that there would be a short across
some of the circuit elements. To avoid this, loosen
the banana plug connectors on the sine wave genera-
tor, and swing away the piece of metal that normally
connects one of the output plugs to the ground.
Tune the sine wave generator’s frequency to reso-
nance, and take the data you’ll need in order to de-
termine x and y at a whole bunch of different places
over one cycle.
Some of the features of the digital oscilloscopes can
make the measurements a lot easier. Doing Acquire>Average
tells the scope to average together a series of up to
128 measurements in order to reduce the amount
of noise. Doing CH 1 MENU>Volts/Div>Fine al-
lows you to scale the display arbitrarily. Rather than
reading voltages by eye from the scope’s x-y grid, you
can make the scope give you a measuring cursor. Do
Cursor>Type>Time. Use the top left knob to move
the cursor to different times. Doing Source>CH 1
and Source>CH 2 gives you the voltage measure-
ment for each channel. (Always use Cursor 1, never
Cursor 2.)
The quality of the results can depend a lot on the
quality of the connections. If the display on the
scope changes noticeably when you wiggle the wires,
you have a problem.

Analysis
Plot y versus x on a piece of graph paper. Let’s
assume that the energy in a field depends on the
field’s strength raised to some power p. Conservation
of energy then gives
|x|p + |y|p = 1 .
Use your graph to determine p, and interpret your
result.

107
35 RC Circuits
Apparatus
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Pasco PI-9587C function generator
1/group unknown capacitor . . . . . . . . . . . . . . . 1/group
known capacitors, 0.05 µF . . . . . . . . . . . . . . . 1/group
resistors of various values

Goals
Observe the exponential curve of a discharging
capacitor.
Determine the capacitance of an unknown ca-
pacitor.

Introduction
God bless the struggling high school math teacher,
but some of them seem to have a talent for mak-
ing interesting and useful ideas seem dull and use-
less. On certain topics such as the exponential func-
tion, ex, the percentage of students who figure out
from their teacher’s explanation what it really means
and why they should care approaches zero. That’s
a shame, because there are so many cases where it’s
useful. The graphs show just a few of the important
situations in which this function shows up.
The credit card example is of the form

y = aet/k ,

while the Chernobyl graph is like

y = ae−t/k ,

In both cases, e is the constant 2.718 . . ., and k is

a positive constant with units of time, referred to
as the time constant. The first type of equation is
referred to as exponential growth, and the second
as exponential decay. The significance of k is that
it tells you how long it takes for y to change by a
factor of e. For instance, an 18% interest rate on closer and closer to zero. For instance, the radioac-
your credit card converts to k = 6.0 years. That tivity near Chernobyl will never ever become exactly
means that if your credit card balance is $1000 in zero. After a while it will just get too small to pose
1996, by 2002 it will be $2718, assuming you never any health risk, and at some later time it will get too
really start paying down the principal. small to measure with practical measuring devices.
An important fact about the exponential function is Why is the exponential function so ubiquitous? Be-
that it never actually becomes zero — it only gets cause it occurs whenever a variable’s rate of change

108 Lab 35 RC Circuits

is proportional to the variable itself. In the credit
card and Chernobyl examples,
(rate of increase of credit card debt)
∝ (current credit card debt)
(rate of decrease of the number of radioactive atoms)
∝ (current number of radioactive atoms)
For the credit card, the proportionality occurs be-
cause your interest payment is proportional to how
much you currently owe. In the case of radioactive
decay, there is a proportionality because fewer re-
maining atoms means fewer atoms available to de-
cay and release radioactive particles. This line of
thought leads to an explanation of what’s so special
about the constant e. If the rate of increase of a vari-
able y is proportional to y, then the time constant
k equals one over the proportionality constant, and
this is true only if the base of the exponential is e,
not 10 or some other number.
Exponential growth or decay can occur in circuits
containing resistors and capacitors. Resistors and
capacitors are the most common, inexpensive, and
simple electrical components. If you open up a cell
phone or a stereo, the vast majority of the parts you
see inside are resistors and capacitors. Indeed, many
useful circuits, known as RC circuits, can be built
out of nothing but resistors and capacitors. In this
lab, you will study the exponential decay of the sim-
plest possible RC circuit, shown below, consisting of
one resistor and one capacitor in series.
Suppose we initially charge up the capacitor, mak-
ing an excess of positive charge on one plate and an
excess of negative on the other. Since a capacitor
behaves like V = Q/C, this creates a voltage dif-
ference across the capacitor, and by Kirchoff’s loop
rule there must be a voltage drop of equal magni-
tude across the resistor. By Ohm’s law, a current
I = V /R = Q/RC will flow through the resistor,
and we have therefore established a proportionality,
(rate of decrease of charge on capacitor)
∝ (current charge on capacitor) .
It follows that the charge on the capacitor will decay
exponentially. Furthermore, since the proportional-
ity constant is 1/RC, we find that the time constant
of the decay equals the product of R and C. (It may
not be immediately obvious that Ohms times Farads
equals seconds, but it does.)
Note that even if we put the charge on the capac-
itor very suddenly, the discharging process still oc-
curs at the same rate, characterized by RC. Thus
RC circuits can be used to filter out rapidly varying
electrical signals while accepting more slowly varying
ones. A classic example occurs in stereo speakers. If
you pull the front panel off of the wooden box that
we refer to as “a speaker,” you will find that there
are actually two speakers inside, a small one for re-
producing high frequencies and a large one for the
low notes. The small one, called the tweeter, not
only cannot produce low frequencies but would ac-
tually be damaged by attempting to accept them.
It therefore has a capacitor wired in series with its
own resistance, forming an RC circuit that filters
out the low frequencies while permitting the highs
to go through. This is known as a high-pass filter.
A slightly different arrangement of resistors and in-
ductors is used to make a low-pass filter to protect
the other speaker, the woofer, from high frequencies.
Observations
In typical filtering applications, the RC time con-
stant is of the same order of magnitude as the pe-
riod of a sound vibration, say ∼ 1 ms. It is therefore
necessary to observe the changing voltages with an
oscilloscope rather than a multimeter. The oscillo-
scope needs a repetitive signal, and it is not possi-
ble for you to insert and remove a battery in the
circuit hundreds of times a second, so you will use
a function generator to produce a voltage that be-
comes positive and negative in a repetitive pattern.
Such a wave pattern is known as a square wave. The
mathematical discussion above referred to the expo-
nential decay of the charge on the capacitor, but an
oscilloscope actually measures voltage, not charge.
As shown in the graphs below, the resulting volt-
age patterns simply look like a chain of exponential
curves strung together.
Make sure that the yellow or red “VAR” knob, on
the front of the knob that selects the time scale, is
109

clicked into place, not in the range where it moves
freely — otherwise the times on the scope are not
calibrated.
A Preliminary observations
Pick a resistor and capacitor with a combined RC
time constant of ∼ 1 ms. Make sure the resistor is
at least ∼ 10kΩ, so that the internal resistance of
the function generator is negligible compared to the
resistance you supply.
Note that the capacitance values printed on the sides
of capacitors often violate the normal SI conventions
about prefixes. If just a number is given on the ca-
pacitor with no units, the implied units are micro-
farads, mF. Units of nF are avoided by the manufac-
turers in favor of fractional microfarads, e.g., instead
of 1 nF, they would use “0.001,” meaning 0.001 µF.
For picofarads, a capital P is used, “PF,” instead of
the standard SI “pF.”
Use the oscilloscope to observe what happens to the
voltages across the resistor and capacitor as the func-
tion generator’s voltage flips back and forth. Note
that the oscilloscope is simply a fancy voltmeter,
so you connect it to the circuit the same way you
would a voltmeter, in parallel with the component
you’re interested in. Make sure the scope is set on
DC, not AC, by doing CH 1>Coupling>DC. A com-
plication is added by the fact that the scope and
the function generator are fussy about having the
grounded sides of their circuits connected to each
other. The banana-to-BNC converter that goes on
the input of the scope has a small tab on one side
marked “GND.” This side of the scope’s circuit must
be connected to the “LO” terminal of the function
generator. This means that when you want to switch
from measuring the capacitor’s voltage to measuring
the resistor’s, you will need to rearrange the circuit
a little.
If the trace on the oscilloscope does not look like the
one shown above, it may be because the function
generator is flip-flopping too rapidly or too slowly.
The function generator’s frequency has no effect on
110 Lab 35 RC Circuits
the RC time constant, which is just a property of
the resistor and the capacitor.
If you think you have a working setup, observe the
effect of temporarily placing a second capacitor in
parallel with the first capacitor. If your setup is
working, the exponential decay on the scope should
become more gradual because you have increased
RC. If you don’t see any effect, it probably means
you’re measuring behavior coming from the internal
R and C of the function generator and the scope.
Use the scope to determine the RC time constant,
and check that it is correct. Rather than reading
times and voltages by eye from the scope’s x-y grid,
you can make the scope give you a measuring cur-
sor. Do Cursor>Type>Time, and Source>CH 1 .
Use the top left knob to move the cursor to different
times.
B Unknown capacitor
Build a similar circuit using your unknown capacitor
plus a known resistor. Use the unknown capacitor
with the same number as your group number. Take
the data you will need in order to determine the RC
time constant, and thus the unknown capacitance.
As a check on your result, obtain a known capacitor
with a value similar to the one you have determined
for your unknown, and see if you get nearly the same
curve on the scope if you replace the unknown ca-
pacitor with the new one.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Plan how you will determine the capacitance
and what data you will need to take.
Analysis
Determine the capacitance, with error analysis (ap-
pendices 2 and 3).
111
36 LRC Circuits
Apparatus
Heath coils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
0.05 µF capacitor . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
Pasco PI-9587C generator (under lab benches in 416)
1/group
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Goals
Observe the resonant behavior of an LRC cir-
cuit.
Observe how the properties of the resonance
curve change when the L, R, and C values are
changed.
Introduction
Radio, TV, cellular phones — it’s mind-boggling
to imagine the maelstrom of electromagnetic waves
that are constantly pass through us and our sur-
roundings. Perhaps equally surprising is the fact
that a radio can pick up a wave with one partic-
ular frequency while rejecting all the others nearly
perfectly. No seasoned cocktail-party veteran could
ever be so successful at tuning out the signals that
are not of interest. What makes radio technology
possible is the phenomenon of resonance, the prop-
erty of an electrical or mechanical system that makes
it respond far more strongly to a driving force that
varies at the same frequency as that at which the de-
vice naturally vibrates. Just as an opera singer can
only break a wineglass by singing the right note, a
radio can be tuned to respond strongly to electrical
forces that oscillate at a particular frequency.
A simplified version of the circuit.
Circuit
As shown in the figure, the circuit consists of the
Heath coil, a 0.05 µF capacitor, a 47-ohm resistor,
112 Lab 36 LRC Circuits
and the sine wave generator to supply a driving volt-
age. You will study the way the circuit resonates,
i.e., responds most strongly to a certain frequency.
Some added complications come from the fact that
the function generator, coil, and oscilloscope do not
behave quite like their idealized versions. The coil
doesn’t act like a pure inductor; it also has a certain
amount of DC resistance, simply because the wire
has finite resistance. So in addition to the 47-ohm
resistor, you will have 62 ohms of resistance coming
from the resistance of the wire in the coil. There is
also some internal resistance from the function gen-
erator itself, amounting to 600 ohms when you use
the outputs marked “high Ω.” The “R” of the circuit
is really the sum of these three series resistances.
You will also want to put the oscilloscope in AC
coupling mode, which filters out any DC component
(additive constant) on the signal. The scope accom-
plishes this filtering by adding in a very small (20
pF) capacitor, which appears in parallel in the cir-
cuit because an oscilloscope, being a voltmeter, is
always used in parallel. In reality, this tiny parallel
capacitance is so small compared to capacitance of
the 0.05 µF capacitor that the resulting correction
is negligible (and that’s a good thing, because if it
wasn’t negligible, the circuit wouldn’t be a simple
series LRC circuit, and its behavior would be much
more complicated).
Observations
A Observation of Resonance
By connecting the oscilloscope to measure the volt-
age across the resistor, you can determine the amount
of power, P = V 2 /R, being taken from the sine
wavegenerator by the circuit and then dissipated as
heat in the resistor. Make sure that your circuit is
hooked up with the resistor connected to the grounded
output of the amplifier, and hook up the oscilloscope
so its grounded connection is on the grounded side
of the resistor. As you change the frequency of the
function generator, you should notice a very strong
response in the circuit centered around one particu-
lar frequency, the resonant frequency fo . (You could
measure the voltage drop across the capacitor or the
inductor instead, but all the pictures of resonance
curves in your textbook are graphs of the behavior
of the resistor. The response curve of a capacitor or
inductor still has a peak at the resonant frequency,
but looks very different off to the sides.)
The inductance of your solenoid is roughly 1 H based
on the approximation that it’s a long, skinny solenoid
(which is not a great approximation here). Based on
this, estimate the resonant frequency of your circuit,
1 time to oscillate many times in between “kicks.” You
ωo = √ . should observe an exponentially decaying sine wave.
LC
Locate ωo accurately, and use it to determine the
inductance of the Heath coil accurately.
B Effect of Changing C
Change the capacitance value by putting two capac-
itors in parallel, and determine the new resonant
frequency. Check whether the resonant frequency
changes as predicted by theory. This is like tuning
your radio to a different frequency. For the rest of
the lab, go back to your original value of C.
C The Width of the Resonance
The width of a resonance is customarily expressed as
the full width at half maximum, ∆f , defined as the
difference in frequency between the two points where
the power dissipation is half of its maximum value.
Determine the FWHM of your resonance. You are
measuring voltage directly, not power, so you need
to find the points where the amplitude of the voltage
across the√resistor drops below its peak value by a
factor of 2.
D Effect of Changing R
Replace the resistor with a 2200-ohm resistor, and
remeasure the FWHM. You should find that the
FWHM has increased in proportion to the resistance.
(Remember that your resistance always includes the
resistance of the coil and the output side of the am-
plifier.)
E Ringing
An LRC circuit will continue oscillating even when
there is no oscillating driving force present. This
unforced behavior is known as “ringing.”Drive your
circuit with a square wave. You can think of this as
if you are giving the circuit repeated “kicks,” so that
it will ring after each kick.
Choose a frequency many many times lower than
the resonant frequency, so that the circuit will have
The rapidity of the exponential decay depends on
how much resistance is in the circuit, since the re-
sistor is the only component that gets rid of energy
permanently. The rapidity of the decay is custom-
arily measured with the quantity Q (for “quality”),
defined as the number of oscillations required for the
potential energy in the circuit to drop by a factor of
535 (the obscure numerical factor being e2π ). For
our purposes, it will be more convenient to extract
Q from the equation
 
πt
Vpeak,i = Vpeak,0 · exp −
QT
where T is the period of the sine wave, Vpeak,0 is the
voltage across the resistor at the peak that we use
to define t = 0, and Vpeak,i is the voltage of a later
peak, occurring at time t.
Collect the data you will need in order to determine
the Q of the circuit, and then do the same for the
other resistance value.
F The Resonance Curve
Going back to your low-resistance setup, collect volt-
age data over a wide range of frequencies, covering
at least a factor of 10 above and below the resonant
frequency. You will want to take closely spaced data
near the resonance peak, where the voltage is chang-
ing rapidly, and less closely spaced points elsewhere.
Far above and far below the resonance, it will be con-
venient just to take data at frequencies that change
by successive factors of two.
(At very high frequencies, above 104 Hz or so, you
may find that rather than continuing to drop off, the
response curve comes back up again. I believe that
this effect arises from nonideal behavior of the coil at
high frequencies: there is stray capacitance between
one loop and the next, and this capacitance acts like
it is in parallel with the coil.)
In engineering work, it is useful to create a graph of
the resonance curve in which the y axis is in decibels,
113
 also find a 6 db/octave slope in the limit of low fre-
  quencies — here the impedance is dominated by the
P capacitor, but the idea is similar. (More complex fil-
db = 10 log10
Pmax tering circuits can achieve roll-offs more drastic than
 
V 6 db/octave.)
= 20 log10 ,
Vmax

and the x axis is a logarithmic frequency scale. (On Prelab

this graph, the FWHM is the width of the curve at 3
db below the peak.) You will construct such a graph The point of the prelab questions is to make sure
from your data. you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
Analysis my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
Check whether the resonant frequency changed by
P1 Using the rough value of L given in the lab
the correct factor when you changed the capacitance.
manual, compute a preliminary estimate of the an-
For both versions of the circuit, compare the FWHM gular frequency ωo , and find the corresponding fre-
of the resonance and the circuit’s Q to the theoretical quency fo .
equations
P2 Express Q in terms of L, R, and C.
R P3 Show that your answer to P2 has the right
∆ω =
L units.
and P4 Using the rough value of L given in the lab
manual, plug numbers into your answer to P2, and
ωo make a preliminary estimate of the Q that you ex-
Q= .
∆ω pect when using the lower of the two resistance val-
Note that there are a total of three resistances in ues. Your result should come out to be 6 (to one sig
series: the 62-ohm resistance of the coil, the 47- fig of precision).
ohm resistor, and the ∼ 50-ohm resistance of the P5 In part D, you could measure t and T using
sine-wave generator’s output. No error analysis is the time scale on the scope. However, all we care
required, since the main errors are systematic ones about is their ratio t/T ; think of a technique for
introduced by the nonideal behavior of the coil and determining t/T that is both more precise and easier
the difficulty of determining an exact, fixed value for to carry out.
the internal resistance of the output of the amplifier.
Graph the resonance curve — you can probably save
yourself a great deal of time by using a computer to
do the calculations and graphing. To do the calcula-
tions, you can go to my web page, www.lightandmatter.com
. Go to the lab manual’s web page, and then click on
“data-analysis tool for the LRC circuits lab”. Once
your data are ready to graph, I suggest using com-
puter software to make your graph (see Appendix
4).
On the high-frequency end, the impedance is dom-
inated by the impedance of the inductor, which is
proportional to frequency. Doubling the frequency
doubles the impedance, thereby cutting the current
by a factor of two and the power dissipated in the re-
sistor by a factor of 4, which is 6.02 db. Since a factor
of 2 in frequency corresponds in musical terms to one
octave, this is referred to as a 6 db/octave roll-off.
Check this prediction against your data. You should

114 Lab 36 LRC Circuits

115
37 Faraday’s Law
Apparatus
function generator . . . . . . . . . . . . . . . . . . . . . . . . 1/group
solenoid (Heath) 1/group plus a few more
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
10-ohm power resistor . . . . . . . . . . . . . . . . . . . . 1/group
4-meter wire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
palm-sized pieces of iron or steel
masking tape
rulers
Goals
Observe electric fields induced by changing mag-
netic fields.
Test Faraday’s law.
Introduction
Physicists hate complication, and when physicist Mi-
chael Faraday was first learning physics in the early
19th century, an embarrassingly complex aspect of
the science was the multiplicity of types of forces.
Friction, normal forces, gravity, electric forces, mag-
netic forces, surface tension — the list went on and
on. Today, 200 years later, ask a physicist to enu-
merate the fundamental forces of nature and the
most likely response will be “four: gravity, electro-
magnetism, the strong nuclear force and the weak
nuclear force.” Part of the simplification came from
the study of matter at the atomic level, which showed
that apparently unrelated forces such as friction, nor-
mal forces, and surface tension were all manifesta-
tions of electrical forces among atoms. The other
big simplification came from Faraday’s experimental
work showing that electric and magnetic forces were
intimately related in previously unexpected ways, so
intimately related in fact that we now refer to the
two sets of force-phenomena under a single term,
“electromagnetism.”
Even before Faraday, Oersted had shown that there
was at least some relationship between electric and
magnetic forces. An electrical current creates a mag-
netic field, and magnetic fields exert forces on an
electrical current. In other words, electric forces
are forces of charges acting on charges, and mag-
netic forces are forces of moving charges on moving
116 Lab 37 Faraday’s Law
charges. (Even the magnetic field of a bar magnet is
due to currents, the currents created by the orbiting
electrons in its atoms.)
Faraday took Oersted’s work a step further, and
showed that the relationship between electricity and
magnetism was even deeper. He showed that a chang-
ing electric field produces a magnetic field, and a
changing magnetic field produces an electric field.
Faraday’s law,
ΓE = −dΦB /dt
relates the circulation of the electric field around a
closed loop to the rate of change of the magnetic
flux through the loop. It forms the basis for such
technologies as the transformer, the electric guitar,
the amplifier, and generator, and the electric motor.
Observations
A Qualitative Observations
To observe Faraday’s law in action you will first need
to produce a varying magnetic field. You can do this
by using a function generator to produce a current
in a solenoid that that varies like a sine wave as a
function of time. The solenoid’s magnetic field will
thus also vary sinusoidally.
The emf in Faraday’s law can be observed around a
loop of wire positioned inside or close to the solenoid.
To make the emf larger and easier to see on an os-
cilloscope, you will use 5-10 loops, which multiplies
the flux by that number of loops.
The only remaining complication is that the rate of
change of the magnetic flux, dΦB /dt, is determined
by the rate of change of the magnetic field, which
relates to the rate of change of the current through
the solenoid, dI/dt. The oscilloscope, however, mea-
sures voltage, not current. You might think that
you could simply observe the voltage being supplied
to the solenoid and divide by the solenoid’s 62-ohm
resistance to find the current through the solenoid.
This will not work, however, because Faraday’s law
produces not only an emf in the loops of wire but also
an emf in the solenoid that produced the magnetic
field in the first place. The current in the solenoid is
being driven not just by the emf from the function
generator but also by this “self-induced” emf. Even
though the solenoid is just a long piece of wire, it
does not obey Ohm’s law under these conditions.
To get around this difficulty, you can insert the 10-
ohm power resistor in the circuit in series with the
function generator and the solenoid. (A power re-
sistor is simply a resistor that can dissipate a large
amount of power without burning up.) The power
resistor does obey Ohm’s law, so by using the scope
to observe the voltage drop across it you can infer
the current flowing through it, which is the same as
the current flowing through the solenoid.
Create the solenoid circuit, and hook up one channel
of the scope to observe the voltage drop across the
power resistor. A sine wave with a frequency on the
order of 1 kHz will work.
Wind the 2-m wire into circular loops small enough
to fit inside the solenoid, and hook it up to the other
channel of the scope.
As always, you need to watch out for ground loops.
The output of the function generator has one of its
terminals grounded, so that ground and the grounded
side of the scope’s input have to be at the same place
in the circuit.
The signals tend to be fairly noise. You can clean
them up a little by having the scope average over a
series of traces. To turn on averaging, do Acquire>Average>128.
To turn it back off, press Sample.
First try putting the loops at the mouth of the solenoid, that, multiply the expression for the field by (cos β +
and observe the emf induced in them. Observe what
happens when you flip the loops over. You will ob-
serve that the two sine waves on the scope are out of
phase with each other. Sketch the phase relationship
in your notebook, and make sure you understand in
terms of Faraday’s law why it is the way it is, i.e.,
why the induced emf has the greatest value at a cer-
tain point, why it is zero at a certain point, etc.
Observe the induced emf at with the loops at several
other positions such as those shown in the figure.
Make sure you understand in the resulting variations
of the strength of the emf in terms of Faraday’s law.
B A Metal Detector
Obtain one of the spare solenoids so that you have
two of them. Substitute it for the loops of wire, so
that you can observe the emf induced in the second
solenoid by the first solenoid. If you put the two
solenoids close together with their mouths a few cm
apart and then insert a piece of iron or steel between
them, you should be able to see a small increase in
the induced emf. The iron distorts the magnetic field
pattern produced by the first solenoid, channeling
more of the field lines through the second solenoid.
C Quantitative Observations
This part of the lab is a quantitative test of Fara-
day’s law. Going back to the setup for part A, mea-
sure the amplitude (peak-to-peak height) of the volt-
age across the power resistor. Choose a position
for the loops of wire that you think will make it
as easy as possible to calculate dΦB /dt accurately
based on knowledge of the variation of the current
in the solenoid as a function of time. Put the loops
in that position, and measure the amplitude of the
induced emf. Repeat these measurements with a fre-
quency that is different by a factor of two.
Self-Check
Before leaving, analyze your results from part C and
make sure you get reasonable agreement with Fara-
day’s law.
Analysis
Describe your observations in parts A and B and
interpret them in terms of Faraday’s law.
Compare your observations in part C quantitatively
with Faraday’s law. The solenoid isn’t very long, so
the approximate expression for the interior field of a
long solenoid isn’t very accurate here. To correct for
cos γ)/2, which you derived in homework problem
11-30 in Simple Nature, where β and γ are angles
between the axis and the lines connecting the point
of interest to the edges of the solenoid’s mouths.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
117
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Plan what raw data you’ll need to collect for
part C, and figure out the equation you’ll use to
test whether your observations are consistent with
Faraday’s law.

118 Lab 37 Faraday’s Law

119
38 Electromagnetism
Apparatus charges. (Even the magnetic field of a bar magnet is
due to currents, the currents created by the orbiting
solenoid (Heath) . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group electrons in its atoms.)
oscilloscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
2-meter wire with banana plugs . . . . . . . . . . . 1/group Faraday took Oersted’s work a step further, and
magnet (stack of 6 Nd) . . . . . . . . . . . . . . . . . . . 1/group showed that the relationship between electricity and
masking tape magnetism was even deeper. He showed that a chang-
string ing electric field produces a magnetic field, and a
changing magnetic field produces an electric field.
Faraday’s work forms the basis for such technologies
as the transformer, the electric guitar, the trans-
Goals former, and generator, and the electric motor. It
also led to the understanding of light as an electro-
Observe electric fields induced by changing mag- magnetic wave.
netic fields.

Build a generator. Qualitative Observations

Discover Lenz’s law.
Introduction
Physicists hate complication, and when physicist Mich- solenoid, producing a current which is detected by
ael Faraday was first learning physics in the early
19th century, an embarrassingly complex aspect of
the science was the multiplicity of types of forces.
Friction, normal forces, gravity, electric forces, mag- as a function of time.
netic forces, surface tension — the list went on and
on. Today, 200 years later, ask a physicist to enu-
merate the fundamental forces of nature and the
most likely response will be “four: gravity, electro-
magnetism, the strong nuclear force and the weak
nuclear force.” Part of the simplification came from
the study of matter at the atomic level, which showed
that apparently unrelated forces such as friction, nor- Do you detect any signal when you move the magnet
mal forces, and surface tension were all manifesta-
tions of electrical forces among atoms. The other
big simplification came from Faraday’s experimental
work showing that electric and magnetic forces were
intimately related in previously unexpected ways, so
intimately related in fact that we now refer to the
two sets of force-phenomena under a single term,
“electromagnetism.”
Even before Faraday, Oersted had shown that there
was at least some relationship between electric and
magnetic forces. An electrical current creates a mag-
netic field, and magnetic fields exert forces on an
electrical current. In other words, electric forces
are forces of charges acting on charges, and mag-
netic forces are forces of moving charges on moving
In this lab you will use a permanent magnet to pro-
duce changing magnetic fields. This causes an elec-
tric field to be induced, which you will detect using
a solenoid (spool of wire) connected to an oscillo-
scope. The electric field drives electrons around the
the oscilloscope. If you haven’t used an oscilloscope
before, your instructor will help you to get started.
It’s simply a device for graphing a measured voltage
A A constant magnetic field
Do you detect any signal on the oscilloscope when
the magnet is simply placed at rest inside the solenoid?
Try the most sensitive voltage scale.
B A changing magnetic field
or wiggle it inside the solenoid or near it? What
happens if you change the speed at which you move
the magnet?
C Moving the solenoid
What happens if you hold the magnet still and move
the solenoid?
The poles of the magnet are its flat faces. In later
parts of the lab you will need to know which is north.
Determine this now by hanging it from a string and
seeing how it aligns itself with the Earth’s field. The
pole that points north is called the north pole of the
magnet. The field pattern funnels into the body of
the magnet through its south pole, and reemerges at

120 Lab 38 Electromagnetism

its north pole.
D A generator
Tape the magnet securely to the eraser end of a pen-
cil so that its flat face (one of its two poles) is like the
head of a hammer, and mark the north and south
poles of the magnet for later reference. Spin the pen-
cil near the solenoid and observe the induced signal.
You have built a generator. (I have unfortunately
not had any luck lighting a lightbulb with the setup,
due to the relatively high internal resistance of the
solenoid.)
Trying Out Your Understanding
E Changing the speed of the generator
If you change the speed at which you spin the pencil,
you will of course cause the induced signal to have a
longer or shorter period. Does it also have any effect
on the amplitude of the wave?
F A solenoid with fewer loops
Use the two-meter cable to make a second solenoid
with the same diameter but fewer loops. Compare
the strength of the induced signals.
G Dependence on distance
How does the signal picked up by your generator
change with distance?
Try to explain what you have observed, and discuss
your interpretations with your instructor.
Lenz’s Law
Lenz’s law describes how the clockwise or counter-
clockwise direction of the induced electric field’s whirlpool
pattern relates to the changing magnetic field. The
main result of this lab is a determination of how
Lenz’s law works. To focus your reasoning, here are
four possible forms for Lenz’s law:
1. The electric field forms a pattern that is clockwise
when viewed along the direction of the B vector of
the changing magnetic field.
2. The electric field forms a pattern that is counter-
clockwise when viewed along the direction of the B
vector of the changing magnetic field.
3. The electric field forms a pattern that is clockwise
when viewed along the direction of the ∆B vector of
the changing magnetic field.
4. The electric field forms a pattern that is coun-
terclockwise when viewed along the direction of the
∆B vector of the changing magnetic field.
Your job is to figure out which is correct.
The most direct way to figure out Lenz’s law is to
chopping motion that ends up with the magnet in
the solenoid, observing whether the pulse induced
is positive or negative. What happens when you
reverse the chopping motion, or when you reverse
the north and south poles of the magnet? Try all
four possible combinations and record your results.
To set up the scope, press DEFAULT SETUP. This
should have the effect of setting the scope on DC
coupling, which is what you want. (If it’s on AC cou-
pling, it tries to filter out any DC part of the input
signals, which distorts the results.) To check that
you’re on DC coupling, you can do CH 1 MENU,
and check that Coupling says DC.
Make sure the scope is on DC coupling, not AC cou-
pling, or your pulses will be distorted.
It can be tricky to make the connection between the
polarity of the signal on the screen of the oscilloscope
and the direction of the electric field pattern. The
figure shows an example of how to interpret a posi-
tive pulse: the current must have flowed through the
scope from the center conductor of the coax cable to
its outer conductor (marked GND on the coax-to-
banana converter).
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 The time-scale for all the signals is determined
121
by the fact that you’re wiggling and waving the mag-
net by hand, so what’s a reasonable order of magni-
tude to choose for the time base on the oscilloscope?

Self-Check
Determine which version of Lenz’s law is correct.

122 Lab 38 Electromagnetism

123
39 Impedance
Observe how the impedances of capacitors and
inductors change with frequency.
Observe how impedances combine according to
the arithmetic of complex numbers.
Setup
We’ll start by observing the impedance of a capaci-
tor. Ideally, what we want is this:
However, we want to know not just the amplitude
of the voltage and current sine-waves but the phase
relationship between them as well, which we can’t
get from a regular meter. We need to use an oscil-
loscope, and oscilloscopes only measure voltage, not
current. This leads us to something like the follow-
ing setup:
Here ch. 2 tells us the voltage across the resistor,
which is related to the current in the resistor accord-
ing to Ohm’s law. By the junction rule, the current
in the resistor is the same as the current through the
capacitor.
But even now, we’re not out of the woods. In this
setup, the ground of ch. 2 is connected to the same
wire as the active (+) connection to ch. 1, which
124 Lab 39 Impedance
would cause ch. 1 to read zero, and would short
across the capacitor as well. Instead, we need this:
Now both GND connections are going to the same
point in the circuit. Because we’ve swapped the con-
nections to ch. 1, its trace will be upside-down, and
inconsistent with ch. 2. There is a special control on
the scope for inverting ch. 2, which makes the two
channels consistent again.
Observations
A Impedance of the capacitor
Hook up the circuit as shown, using a 1 kΩ resistance
and a 0.2 µF capacitance. The HP signal genera-
tor has a ground strap connecting one of its output
terminals to ground. Disconnect this ground strap,
since grounding either side of the signal generator
would mean that either the resistor or the capacitor
would be connected to ground on both sides. Try a
frequency of 100 Hz.
Observe the phase relationship between VC , on ch.
1, and the signal on ch. 2, which essentially tells
us the current IC except for a factor of 1/R. Sketch
this phase relationship in your raw data. Because
VC = q/C and I = dq/dt, the current through the
capacitor should be proportional to dV/dt. Based
on the phase relationship you observed, does this
seem to be true?
Measure the phase angle numerically from the oscil-
loscope. Is it what you expect?
Determine the magnitude of the capacitor’s impedance.
Suppose you represent the signal that is ahead in
phase using a point that is more counterclockwise
in the complex plane. Sketch the locations of the
voltage and current in the complex plane. (You can
arbitrarily choose one of them to be along the real
axis if you like.) Where would the impedance then
lie in the plane?
Now change the frequency to 1000 Hz, and see what
changes. Sketch your new impedance in the com-
plex plane. Do you find the expected relationship
between impedance and frequency?

B Inductance of the Heath coil

Make the measurements you need in order to calcu-
late the theoretical inductance of the inductor, using
the equation derived in the prelab. The approxima-
tion may be off by as much as a factor of two, since
the solenoid isn’t long and skinny, but it’s useful so
you have some idea of what to expect.

C Impedance of the inductor

Now repeat all the above steps using the Heath coil
as an inductor.

D Impedances in series
Put the capacitor and inductor in series, and collect
the data you’ll need in order to determine their com-
bined impedance at several frequencies ranging from
100 to 1000 Hz.

Analysis
Use your data from part C to determine an experi-
mental value of the coil’s inductance, and compare
with the theoretical result based on your measure-
ments in part B.
Graph the theoretical and experimental impedance
of the series combination in part D, overlaying them
on the same graph. Show theory as a curve and ex-
periment as discrete data-points. Do the same kind
of graph for the parallel combination.

125
40 Refraction and Images
Apparatus
rectangular block of plastic (20x10x5 cm,
from blackboard optics kit), or plastic box with wa-
ter in it
laser
spiral plastic tube and fiber optic cable for demon-
strating total internal reflection
ruler
protractor
butcher paper
Goals
Observe the phenomena of refraction and total
internal reflection.
Locate a virtual image in a plastic block by
ray tracing, and compare with the theoretically
predicted position of the image.
Introduction
Without the phenomenon of refraction, the lens of
your eye could not focus light on your retina, and you
would not be able to see. Refraction is the bending of
rays of light that occurs when they pass through the
boundary between two media in which the speed of
light is different. Light entering your eye passes from
air, in which the speed of light is 3.0 × 108 m/s, into
the watery tissues of your eye, in which it is about
2.2 × 108 m/s. Since it is inconvenient to write or
say the speed of light in a particular medium, we
usually speak in terms of the index of refraction, n,
defined by
n = c/v,
where c is the speed of light in a vacuum, and v is
the speed of light in the medium in question. Thus,
vacuum has n = 1 by definition. Air, which is not
very dense, does not slow light down very much, so
it has an index of refraction very close to 1. Water
has an index of refraction of about 1.3, meaning that
light moves more slowly in water by a factor of 1/1.3.
Refraction, the bending of light, occurs for the fol-
lowing reason. Imagine, for example, a beam of light
entering a swimming pool at an angle. Because of
the angle, one side of the beam hits the water first,
126 Lab 40 Refraction and Images
and is slowed down. The other side of the beam,
however, gets to travel in air, at its faster speed, for
longer, because it enters the water later — by the
time it enters the water, the other side of the beam
has been limping along through the water for a little
while, and has not gotten as far. The wavefront is
therefore twisted around a little, in the same way
that a marching band turns by having the people on
one side take smaller steps.
Quantitatively, the amount of bending is given by
Snell’s law:
ni sin θi = nt sin θt ,
where the index i refers to the incident light and in-
cident medium, and t refers to the transmitted light
and the transmitting medium. Note that the an-
gles are defined with respect to the normal, i.e., the
imaginary line perpendicular to the boundary.
Also, not all of the light is transmitted. Some is re-
flected — the amount depends on the angles. In fact,
for certain values of ni , nt , and θi , there is no value
of θt that will obey Snell’s law (sin θt would have
to be greater than one). In such a situation, 100%
of the light must be reflected. This phenomenon is
known as total internal reflection. The word inter-
nal is used because the phenomenon only occurs for
ni > nt . If one medium is air and the other is plastic
or glass, then this can only happen when the incident
light is in the plastic or glass, i.e., the light is try-
ing to escape but can’t. Total internal reflection is
used to good advantage in fiber-optic cables used to
transmit long-distance phone calls or data on the in-
ternet — light traveling down the cable cannot leak
out, assuming it is initially aimed at an angle close
enough to the axis of the cable.
Although most of the practical applications of the
phenomenon of refraction involve lenses, which have
curved shapes, in this lab you will be dealing almost
exclusively with flat surfaces.

Preliminaries

Check whether your laser’s beam seems to be roughly

parallel.

Observations
A Index of refraction of plastic
Make the measurements you have planned in order
to determine the index of refraction of the plastic
block (or the water, whichever you have). The laser
and the block of plastic can simply be laid flat on the
table. Make sure that the laser is pointing towards
the wall.

B Total internal reflection

Try shining the laser into one end of the spiral-
shaped plastic rod. If you aim it nearly along the
axis of the cable, none will leak out, and if you put
your hand in front of the other end of the rod, you
will see the light coming out the other end. (It will
not be a well-collimated beam any more because the
beam is spread out and distorted when it undergoes
the many reflections on the rough and curved inside
the rod.)
There’s no data to take. The point of having this as
part of the lab is simply that it’s hard to demonstrate
to a whole class all at once.

C A virtual image
Pick up the block, and have your partner look side-
ways through it at your finger, touching the sur-
face of the block. Have your partner hold her own
finger next to the block, and move it around un-
til it appears to be as far away as your own finger.
Her brain achieves a perception of depth by subcon-
sciously comparing the images it receives from her
two eyes. Your partner doesn’t actually need to be extrapolate the rays leaving the block back into the
able to see her own finger, because her brain knows block. They should all appear to have come from the
how to position her arm at a certain point in space. same point, where you saw the virtual image. You’ll
Measure the distance di , which is the depth of the need to photocopy the tracing so that each person
image of your finger relative to the front of the block. can turn in a copy with his or her writeup.
Now trace the outline of the block on a piece of pa-
per, remove the block, mark the location of the im- Prelab
age, and put the block back on the paper. Shine
the laser at the point where your finger was origi- The point of the prelab questions is to make sure
nally touching the block, observe the refracted beam, you understand what you’re doing, why you’re do-
and draw it in. Repeat this whole procedure several ing it, and how to avoid some common mistakes. If
times, with the laser at a variety of angles. Finally, you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise

127
you’re just setting yourself up for failure in lab.
Do the laser safety checklist, Appendix 9, tear it out,
and turn it in at the beginning of lab. If you don’t
understand something, don’t initial that point, and
ask your instructor for clarification before you start
the lab.
P1 Laser beams are supposed to be very nearly
parallel (not spreading out or contracting to a focal
point). Think of a way to test, roughly, whether this
is true for your laser.
P2 Plan how you will determine the index of re-
fraction in part A.
P3 You have complete freedom to choose any in-
cident angle you like in part A. Discuss what choice
would give the highest possible precision for the mea-
surement of the index of refraction.

Analysis
Using your data for part A, extract the index of re-
fraction. Estimate the accuracy of your raw data,
and determine error bars for your index of refrac-
tion.
Using trigonometry and Snell’s law, make a the-
oretical calculation of di . You’ll need to use the
small-angle approximation sin θ ≈ tan θ ≈ θ, for θ
measured in units of radians. (For large angles, i.e.
viewing the finger from way off to one side, the rays
will not converge very closely to form a clear virtual
image.)
Explain your results in part C and their meaning.
Compare your three values for di : the experimental
value based on depth perception, the experimental
value found by ray-tracing with the laser, and the
theoretical value found by trigonometry.

128 Lab 40 Refraction and Images

129
41 Geometric Optics
Apparatus
optical bench . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
converging lens (unknown focal
length to be measured) . . . . . . . . . . . . . . . . . . . 1/group
converging lens, longest available
focal length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1/group
converging lens, f = 50 mm
1/group lamp and arrow-shaped mask . . . . 1/group
frosted glass screen . . . . . . . . . . . . . . . . . . . . . . . 1/group
Goals
Observe a real image formed by a convex lens,
and determine its focal length.
Construct a telescope and measure its angular
magnification.
Introduction
The credit for invention of the telescope is disputed,
but Galileo was probably the first person to use one
for astronomy. He first heard of the new invention
when a foreigner visited the court of his royal pa-
trons and attempted to sell it for an exorbitant price.
Hearing through second-hand reports that it con-
sisted of two lenses, Galileo sent an urgent message
to his benefactors not to buy it, and proceeded to
reproduce the device himself. An early advocate of
simple scientific terminology, he wanted the instru-
ment to be called the “occhialini,” Italian for “eye-
thing,” rather than the Greek “telescope.”
His astronomical observations soon poked some gap-
ing holes in the accepted Aristotelian view of the
heavens. Contrary to Aristotle’s assertion that the
heavenly bodies were perfect and without blemishes,
he found that the moon had mountains and the sun
had spots (the marks on the moon visible to the
naked eye had been explained as optical illusions or
atmospheric phenomena). This put the heavens on
an equal footing with earthly objects, paving the
way for physical theories that would apply to the
whole universe, and specifically for Newton’s law of
gravity. He also discovered the four largest moons
of Jupiter, and demonstrated his political savvy by
naming them the “Medicean satellites” after the pow-
erful Medici family. The fact that they revolved
130 Lab 41 Geometric Optics
around Jupiter rather than the earth helped make
more plausible Copernicus’ theory that the planets
did not revolve around the earth but around the sun.
Galileo’s ideas were considered subversive, and many
people refused to look through his telescope, either
because they thought it was an illusion or simply
because it was supposed to show things that were
contrary to Aristotle.
The figure on the next page shows the simplest re-
fracting telescope. The object is assumed to be at
infinity, so a real image is formed at a distance from
the objective lens equal to its focal length, fo . By
setting up the eyepiece at a distance from the image
equal to its own focal length, fE , light rays that were
parallel are again made parallel.
The point of the whole arrangement is angular mag-
nification. The small angle θ1 is converted to a large
θ2 . It is the small angular size of distant objects that
makes them hard to see, not their distance. There is
no way to tell visually whether an object is a thirty
meters away or thirty billion. (For objects within a
few meters, your brain-eye system gives you a sense
of depth based on parallax.) The Pleiades star clus-
ter can be seen more easily across many light years
than Mick Jagger’s aging lips across a stadium. Peo-
ple who say the flying saucer “looked as big as an
aircraft carrier” or that the moon “looks as big as
a house” don’t know what they’re talking about.
The telescope does not make things “seem closer”
— since the rays coming at your eye are parallel,
the final virtual image you see is at infinity. The
angular magnification is given by
MA = θ2 /θ1
(to be measured directly in this lab)
MA = fo /fE
(theory)
Observations
A Focal length of a convex lens
Use your unknown convex lens to project a real im-
age on the frosted glass screen. For your object, use
the lamp with the arrow-shaped aperture in front of
it. Make sure to lock down the parts on the opti-
cal bench, or else they may tip over and break the
optics!
B The telescope greatest possible magnification.
Use your optical bench and your two known lenses
to build a telescope. Since the telescope is a device
for viewing objects at infinity, you’ll want to take it
Analysis
outside. Determine the focal length of the unknown lens, with
The best method for determining the angular magni- error bars.
fication is to observe the same object with both eyes Find the angular magnification of your telescope from
open, with one eye looking through the telescope and your data, with error bars, and compare with the-
one seeing the object without the telescope. Good ory. Do they agree to within the accuracy of the
precision can be obtained, for example, by looking at measurement?
a large object like a coke machine, and determining
that a small part of it, whose size you can measure
with a ruler, appears, when magnified, to cover some
larger part of it, which you can also measure.
Your brain is not capable of focusing one eye at one
distance, and the other at another distance. There-
fore it’s important to get your telescope adjusted
precisely so that the image is at infinity. You can do
this by focusing your naked eye on a distant object,
and then moving the objective until the image pops
into focus in the other eye. Theoretically this would
be accomplished simply by setting the lenses at the
distance shown in the diagram, but in reality, a small
amount of further adjustment is necessary, perhaps
because the quality control on the focal lengths of
the lenses is not perfect.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
Do the laser safety checklist, Appendix 9, tear it out,
and turn it in at the beginning of lab. If you don’t
understand something, don’t initial that point, and
ask your instructor for clarification before you start
the lab.
P1 In part A, do you want the object to be closer
to the lens than the lens’ focal length, exactly at a
distance of one focal length, or farther than the focal
length? What about the screen?
P2 Plan what measurements you will make in part
A and how you will use them to determine the lens’
focal length.
P3 It’s disappointing to construct a telescope with
a very small magnification. Given a selection of
lenses, plan how you can make a telescope with the

131
A refracting telescope

132 Lab 41 Geometric Optics

133
42 Two-Source Interference
Apparatus
ripple tank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
yellow foam pads . . . . . . . . . . . . . . . . . . . . . . . . . 4/group
lamp and unfrosted straight-filament bulb
1/group wave generator . . . . . . . . . . . . . . . . . . .1/group
big metal L-shaped arms for hanging
the wave generator . . . . . . . . . . . . . . . . . . . . . . . 1/group
little metal L-shaped arms with yellow
plastic balls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
rubber bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
white plastic screen . . . . . . . . . . . . . . . . . . . . . . . 1/group
Thornton DC voltage source . . . . . . . . . . . . . . 1/group
small rubber stopper . . . . . . . . . . . . . . . . . . . . . 1/group
power strip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
bucket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
mop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
flathead screwdriver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
rulers and protractors
kimwipes and alcohol for cleaning
butcher paper
important ideas about light as a wave can be seen
in one simple experiment, shown in the first figure.1
A wave comes up from the bottom of the page, and
encounters a wall with two slits chopped out of it.
The result is a fan pattern, with strong wave motion
coming out along directions like X and Z, but no
vibration of the water at all along lines like Y. The
reason for this pattern is shown in the second figure.
The two parts of the wave that get through the slits
create an overlapping pattern of ripples. To get to
a point on line X, both waves have to go the same
distance, so they’re in step with each other, and re-
inforce. But at a point on line Y, due to the unequal
distances involved, one wave is going up while the
other wave is going down, so there is cancellation.
The angular spacing of the fan pattern depends on
both the wavelength of the waves, λ, and the dis-
tance between the slits, d.

Goals
Observe how a 2-source interference pattern of
water waves depends on the distance between
the sources. The ripple tank is tank that sits about 30 cm above
the floor. You put a little water in the tank, and
produce waves. There is a lamp above it that makes
Observations a point-like source of light, and the waves cast pat-
terns of light on a screen placed on the floor. The
Light is really made of waves, not rays, so when we patterns of light on the screen are easier to see and
treated it as rays, we were making an approximation. measure than the ripples themselves.
You might think that when the time came to treat
light as a wave, things would get very difficult, and In reality, it’s not very convenient to produce a double-
it would be hard to predict or understand anything slit diffraction pattern exactly as depicted in the first
without doing complicated calculations. figure, because the waves beyond the slits are so
weak that they are difficult to observe clearly. In-
stead, you’ll simply produce synchronized circular
ripples from two sources driven by a motor.
Put the tank on the floor. Plug the hole in the side of
the tank with the black rubber stopper. If the plastic
is dirty, clean it off with alcohol and kimwipes. Wet
the four yellow foam pads, and place them around
the sides of the tank. Pour in water to a depth of
about 5-7 mm. Adjust the metal feet to level the
tank, so that the water is of equal depth throughout
1 The photo is from the textbook PSSC physics, which has

Life isn’t that bad. It turns out that all of the most a blanket permission for free use after 1970.

134 Lab 42 Two-Source Interference

the tank. (Do not rotate the wooden legs them-
selves, just the feet.) If too many bubbles form on
the plastic, wipe them off with a ruler.
Make sure the straight-filament bulb in the light
source is rotated so that when you look in through
the hole, you are looking along the length of the fil-
ament. This way the lamp acts like a point source
of light above the tank. To test that it’s oriented
correctly, check that you can cast a perfectly sharp
image of the tip of a pen.
Clamp the light source to the post and turn it on.
Put the white plastic screen on the floor under the
tank. If you make ripples in the water, you should
be able to see the wave pattern on the screen.
The wave generator consists of a piece of wood that
hangs by rubber bands from the two L-shaped metal
hangers. There is a DC motor attached, which spins
an intentionally unbalanced wheel, resulting in vi-
bration of the wood. The wood itself can be used
to make straight waves directly in the water, but
in this experiment you’ll be using the two little L-
shaped pieces of metal with the yellow balls on the
end to make two sources of circular ripples. The DC
motor runs off of the DC voltage source, and the
more voltage you supply, the faster the motor runs.
Start just by sticking one little L-shaped arm in the
piece of wood, and observing the circular wave pat-
tern it makes. Now try two sources at once, in neigh-
boring holes. Pick a speed (frequency) for the motor
that you’ll use throughout the experiment — a fairly
low speed works well. Measure the angular spacing
of the resulting diffraction pattern for several values
of the spacing, d, between the two sources of ripples.
How do you think the angular spacing of the wave
pattern seems to depend mathematically on d? Con-
struct a graph to test whether this was really true.
If you’re not sure what mathematical rule to guess,
you can use the methods explained in Appendix 5
and look for any kind of a power law relationship.

135
43 Wave Optics
Apparatus
helium-neon laser
1/group optical bench with posts & holders 1/group
high-precision double slits . . . . . . . . . . . . . . . . 1/group
rulers
meter sticks
tape measures
butcher paper
Goals
Observe evidence for the wave nature of light.
Determine the wavelength of the red light emit-
ted by your laser, by measuring a double-slit
diffraction pattern. (The part of the spectrum
that appears red to the human eye covers quite
a large range of wavelengths. A given type of
laser, e.g., He-Ne or solid-state, will produce
one very specific wavelength.)
Determine the approximate diameter of a hu-
man hair, using its diffraction pattern.
Introduction
Isaac Newton’s epitaph, written by Alexander Pope,
reads:
Nature and Nature’s laws lay hid in night.
God said let Newton be, and all was light.
Notwithstanding Newton’s stature as the greatest
physical scientist who ever lived, it’s a little ironic
that Pope chose light as a metaphor, because it was
in the study of light that Newton made some of his
worst mistakes. Newton was a firm believer in the
dogma, then unsupported by observation, that mat-
ter was composed of atoms, and it seemed logical to
him that light as well should be composed of tiny
particles, or “corpuscles.” His opinions on the sub-
ject were so strong that he influenced generations
of his successors to discount the arguments of Huy-
gens and Grimaldi for the wave nature of light. It
was not until 150 years later that Thomas Young
demonstrated conclusively that light was a wave.
Young’s experiment was incredibly simple, and could
probably have been done in ancient times if some
136 Lab 43 Wave Optics
savvy Greek or Chinese philosopher had only thought
of it. He simply let sunlight through a pinhole in a
window shade, forming what we would now call a
coherent beam of light (that is, a beam consisting
of plane waves marching in step). Then he held a
thin card edge-on to the beam, observed a diffrac-
tion pattern on a wall, and correctly inferred the
wave nature and wavelength of light. Since Roemer
had already measured the speed of light, Young was
also able to determine the frequency of oscillation of
the light.
Today, with the advent of the laser, the production
of a bright and coherent beam of light has become
as simple as flipping a switch, and the wave nature
of light can be demonstrated very easily. In this lab,
you will carry out observations similar to Young’s,
but with the benefit of hindsight and modern equip-
ment.
Observations
A Determination of the wavelength of red light
Set up your laser on your optical bench. You will
want as much space as possible between the laser
and the wall, in order to let the diffraction pattern
spread out as much as possible and reveal its fine
details.
Tear off two small scraps of paper with straight edges.
Hold them close together so they form a single slit.
Hold this improvised single-slit grating in the laser
beam and try to get a single-slit diffraction pattern.
You may have to play around with different widths
for the slit. No quantitative data are required. This
is just to familiarize you with single-slit diffraction.
Make a diffraction pattern with the double-slit grat-
ing. See what happens when you hold it in your
hand and rotate it around the axis of the beam.
The diffraction pattern of the double-slit grating con-
sists of a rapidly varying pattern of bright and dark
bars, with a more slowly varying pattern superim-
posed on top. (See the figure two pages after this
page.) The rapidly varying pattern is the one that
is numerically related to the wavelength, λ, and the
distance between the slits, d, by the equation
∆θ = λ/d,
where θ is measured in radians. To make sure you
can see the fine spacing, put your slits several meters Prelab
away from the wall. This will necessitate shining it
across the space between lab tables. To make it less The point of the prelab questions is to make sure
likely that someone will walk through the beam and you understand what you’re doing, why you’re do-
get the beam in their eye, put some of the small ing it, and how to avoid some common mistakes. If
desks under the beam. The slit patterns we’re using you don’t know the answers, make sure to come to
actually have three sets of slits, with the following my office hours before lab and get help! Otherwise
dimensions: you’re just setting yourself up for failure in lab.
w (mm) d (mm)
Read the safety checklist.
A .12 .6
B .24 .6 P1 Roughly what wavelength do you expect red
C .24 1.2 light to have?
The small value of d is typically better, for two rea-
P2 It is not practical to measure ∆θ directly us-
sons: (1) it produces a wider diffraction pattern,
ing a protractor. Plan how you will determine ∆θ
which is easier to see; (2) it’s easy to get the beam of
indirectly, via trigonometry.
the laser to cover both slits. If your diffraction pat-
tern doesn’t look like the one in the figure on page P3 Make a rough order-of-magnitude guess of the
136, typically the reason is that you’re only cover- diameter of a human hair.
ing one slit with the beam (in which case you get a
single-slit diffraction pattern), or you’re not illumi-
nating the two slits equally (giving a funny-looking Analysis
pattern with little dog-bones and things in it).
Determine the wavelength of the light and the diam-
Think about the best way to measure the spacing of eter of the hair, with error bars.
the pattern accurately. Is it best to measure from a
bright part to another bright part, or from dark to
dark? Is it best to measure a single spacing, or take
several spacings and divide by the number to find
what one spacing is? Do it.
Determine the wavelength of the light, in units of
nanometers. Make sure it is in the right range for
red light.

B Diameter of a human hair

Pull out one of your own hairs, hold it in the laser
beam, and observe a diffraction pattern. It turns
out that the diffraction pattern caused by a narrow
obstruction, such as your hair, has the same spac-
ing as the pattern that would be created by a sin-
gle slit whose width was the same as the diameter
of your hair. (This is an example of a general theo-
rem called Babinet’s principle.) Measure the spacing
of the diffraction pattern. (Since the hair’s diame-
ter is the only dimension involved, there is only one
diffraction pattern with one spacing, not superim-
posed fine and coarse patterns as in part A.) De-
termine the diameter of your hair. Make sure the
value you get is reasonable, and compare with the
order-of-magnitude guess you made in your prelab
writeup.

137
A double-slit diffraction pattern.

138 Lab 43 Wave Optics

139
44 Polarization
Apparatus
laser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
calcite crystal (flattest available) . . . . . . . . . . 1/group
polarizing films . . . . . . . . . . . . . . . . . . . . . . . . . . . 2/group
Na gas discharge tube . . . . . . . . . . . . . . . . . . . . 1/group
photovoltaic cell and collimator . . . . . . . . . . . 1/group
Goals
Make qualitative observations about the polar-
ization of light.
Test quantitatively the hypothesis that polar-
ization relates to the direction of the field vec-
tors in an electromagnetic wave.
Introduction
It’s common knowledge that there’s more to light
than meets the eye: everyone has heard of infrared
and ultraviolet light, which are visible to some other
animals but not to us. Another invisible feature of
the wave nature of light is far less well known. Elec-
tromagnetic waves are transverse, i.e., the electric
and magnetic field vectors vibrate in directions per-
pendicular to the direction of motion of the wave.
Two electromagnetic waves with the same wavelength
can therefore be physically distinguishable, if their
electric and magnetic fields are twisted around in
different directions. Waves that differ in this way
are said to have different polarizations.
An electromagnetic wave has electric and magnetic field
vectors that vibrate in the directions perpendicular to its
direction of motion. The wave’s direction of polarization is
defined as the line along which the electric field lies.
Maybe we polarization-blind humans are missing out
on something. Some insects and crustaceans can de-
140 Lab 44 Polarization
tect polarization, and a neuroscientist at the Univer-
sity of Pennsylvania has recently found evidence that
a freshwater fish called the green sunfish can see the
polarization of light (Discover magazine, Oct. 1996).
Most sources of visible light (such as the sun or a
light bulb) are unpolarized. An unpolarized beam
of light contains a random mixture of waves with
many different directions of polarization, all of them
changing from moment to moment, and from point
to point within the beam.
Qualitative Observations
Before doing anything else, turn on your gas dis-
charge tube, so it will be warmed up when you are
ready to do part E.
A Double refraction in calcite
Place a calcite crystal on this page. You will see two
images of the print through the crystal.
To understand why this happens, try shining the
laser beam on a piece of paper and then inserting
the calcite crystal in the beam. If you rotate the
crystal around in different directions, you should be
able to get two distinct spots to show up on the
paper. (This may take a little trial and error, partly
because the effect depends on the correct orientation
of the crystal, but also because the crystals are not
perfect, and it can be hard to find a nice smooth
spot through which to shine the beam.)
In the refraction lab, you’ve already seen how a beam
of light can be bent as it passes through the interface
between two media. The present situation is similar
because the laser beam passes in through one face of
the crystal and then emerges from a parallel face at
the back. You have already seen that in this type of
situation, when the beam emerges again, its direc-
tion is bent back parallel to its original direction, but
the beam is offset a little bit. What is different here
is that the same laser beam splits up into two parts,
which bumped off course by different amounts.
What’s happening is that calcite, unlike most sub-
stances, has a different index of refraction depending
on the polarization of the light. Light travels at a
different speed through calcite depending on how the
electric and magnetic fields are oriented compared to
the crystal. The atoms inside the crystal are packed
in a three-dimensional pattern sort of like a stack of
oranges or cannonballs. This packing arrangement
has a special axis of symmetry, and light polarized
along that axis moves at one speed, while light polar-
ized perpendicular to that axis moves at a different
speed.
It makes sense that if the original laser beam was
a random mixture of all possible directions of po-
larization, then each part would be refracted by a
different amount. What is a little more surprising is
that two separated beams emerge, with nothing in
between. The incoming light was composed of light
with every possible direction of polarization. You
would therefore expect that the part of the incoming
light polarized at, say, 45 ◦ compared to the crystal’s
axis would be refracted by an intermediate amount,
but that doesn’t happen. This surprising observa-
tion, and all other polarization phenomena, can be
understood based on the vector nature of electric
and magnetic fields, and the purpose of this lab is
to lead you through a series of observations to help
you understand what’s really going on.
B A polarized beam entering the calcite
A single laser beam entering a calcite crystal breaks up
into two parts, which are refracted by different amounts.
The calcite splits the wave into two parts, polarized in
perpendicular directions compared to each other.
We need not be restricted to speculation about what
was happening to the part of the light that entered
the calcite crystal polarized at a 45 ◦ angle. You can
use a polarizing film, often referred to informally as a
“Polaroid,” to change unpolarized light into a beam
of only one specific polarization. In this part of the
lab, you will use a polarizing film to produce a beam
of light polarized at a 45 ◦ angle to the crystal’s in-
ternal axis.
If you simply look through the film, it doesn’t look
like anything special — everything just looks dim-
mer, like looking through sunglasses. The light reach-
ing your eye is polarized, but your eye can’t tell that.
If you looked at the film under a microscope, you’d
see a pattern of stripes, which select only one direc-
tion of polarization of the light that passes through.
Now try interposing the film between the laser and
the crystal. The beam reaching the crystal is now
polarized along some specific direction. If you rotate
the film, you change beam’s direction of polariza-
tion. If you try various orientations, you will be able
to find one that makes one of the spots disappear,
and another orientation of the film, at a 90 ◦ angle
compared to the first, that makes the other spot go
away. When you hold the film in one of these direc-
tions, you are sending a beam into the crystal that
is either purely polarized along the crystal’s axis or
purely polarized at 90 ◦ to the axis.
By now you have already seen what happens if the
film is at an intermediate angle such as 45 ◦ . Two
spots appear on the paper in the same places pro-
duced by an unpolarized source of light, not just a
single spot at the midpoint. This shows that the
crystal is not just throwing away the parts of the
light that are out of alignment with its axis. What
is happening instead is that the crystal will accept a
beam of light with any polarization whatsoever, and
split it into two beams polarized at 0 and 90 ◦ compared
to the crystal’s axis.
This behavior actually makes sense in terms of the
wave theory of light. Light waves are supposed to
obey the principle of superposition, which says that
waves that pass through each other add on to each
other. A light wave is made of electric and magnetic
fields, which are vectors, so it is vector addition we’re
141
talking about in this case. A vector at a 45 ◦ angle
can be produced by adding two perpendicular vec-
tors of equal length. The crystal therefore cannot
respond any differently to 45-degree polarized light
than it would to a 50-50 mixture of light with 0-
degree and 90-degree polarization.
The principle of superposition implies that if the 0 ◦ and
90 ◦ polarizations produce two different spots, then the
two waves superimposed must produce those two spots,
not a single spot at an intermediate location.
C Two polarizing films
So far I’ve just described the polarizing film as a
device for producing polarized light. But one can
apply to the polarizing film the same logic of super-
position and vector addition that worked with the
calcite crystal. It would not make sense for the film
simply to throw away any waves that were not per-
fectly aligned with it, because a field oriented on a
slant can be analyzed into two vector components,
at 0 and 90 ◦ with respect to the film. Even if one
component is entirely absorbed, the other compo-
nent should still be transmitted.
Based on these considerations, now think about what
will happen if you look through two polarizing films
at an angle to each other, as shown in the figure
above. Do not look into the laser beam! Just look
around the room. What will happen as you change
the angle θ?
D Three polarizing films
Now suppose you start with two films at a 90 ◦ angle
to each other, and then sandwich a third film be-
tween them at a 45 ◦ angle, as shown in the two fig-
142 Lab 44 Polarization
ures above. Make a prediction about what will hap-
pen, and discuss your prediction with your instructor
before you make the actual observation.
Quantitative Observations
E Intensity of light passing through two polar-
izing films
In this part of the lab, you will make numerical mea-
surements of the transmission of initially unpolarized
light transmitted through two polarizing films at an
angle θ to each other. To measure the intensity of
the light that gets through, you will use a photocell,
which is a device that converts light energy into an
electric current.
You will use a voltmeter to measure the voltage
across the photocell when light is shining on it. A
photovoltaic cell is a complicated nonlinear device,
but I’ve found empirically that under the conditions
we’re using in this experiment, the voltage is pro-
portional to the power of the light striking the cell:
twice as much light results in twice the voltage.
This measurement requires a source of light that is
unpolarized, constant in intensity, and comes from
a specific direction so it can’t get to the photocell
without going through the polaroids. The ambient
light in the room is nearly unpolarized, but varies
randomly as people walk in front of the light fixtures,
etc. The laser beam is constant in intensity, but as
I was creating this lab I found to my surprise that it
is partially polarized, with a polarization that varies
over time. A more suitable source of light is the
sodium gas discharge tube, which makes a nearly
monochromatic, unpolarized yellow light. Make sure
you have allowed it to warm up for at least 15-20
minutes before using it; before it warms up, it makes
a reddish light, and the polaroids do not work very
well on that color.
Make measurements of the relative intensity of light
transmitted through the two polarizing films, using a
variety of angles θ. Don’t assume that the notches on
the plastic housing of the polarizing films are a good
indication of the orientation of the films themselves.

Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
P1 Given the angle θ between the polarizing films,
predict the ratio |E0 |/|E| of the transmitted electric
field to the incident electric field.
P2 Based on your answer to P1, predict the ra-
tio P 0 /P of the transmitted power to the incident
power.
P3 Sketch a graph of your answer to P2. Super-
imposed on the same graph, show a qualitative pre-
diction of how it would change if the polaroids were
not 100% perfect at filtering out one component of
the field.

Analysis
Discuss your qualitative results in terms of superpo-
sition and vector addition.
Graph your results from part E, and superimpose a
theoretical curve for comparison. Discuss how your
results compare with theory. Since your measure-
ments of light intensity are relative, just scale the
theoretical curve so that its maximum matches that
of the experimental data. (You might think of com-
paring the intensity transmitted through the two po-
laroids with the intensity that you get with no po-
laroids in the way at all. This doesn’t really work,
however, because in addition to acting as polarizers,
the polaroids simply absorb a certain percentage of
the light, just as any transparent material would.)

143
45 The Photoelectric Effect
Apparatus of light could be visualized as a stream of machine-
gun bullets. The electrons would be small targets,
Hg gas discharge tube but when a “light bullet” did score a hit, it packed
light aperture assembly enough of an individual wallop to knock the elec-
lens/grating assembly tron out immediately. Based on other experiments
photodiode module, support base, and coupling rod involving the spectrum of light emitted by hot, glow-
digital multimeter (Fluke) ing objects, Einstein also proposed that each photon
pieces of plywood had an energy given by
green and yellow filters
E = hf ,

where f is the frequency of the light and h is Planck’s

Goals constant.
Observe evidence that light has particle prop-
In this lab, you will perform the classic experiment
erties as well as wave properties.
used to test Einstein’s theory. You should refer to
the description of the experiment in your textbook.
Measure Planck’s constant.
Briefly, you will expose the metal cathode of a vac-
uum tube to light of various frequencies, and deter-
mine the voltage applied between the cathode and
Introduction anode that just barely suffices to cut off the pho-
The photoelectric effect, a phenomenon in which toelectric current completely. This is known as the
light shakes an electron loose from an object, pro- stopping voltage, Vs . According to Einstein’s theory,
vided the first evidence for wave-particle duality: the stopping voltage should obey the equation
the idea that the basic building blocks of light and
matter show a strange mixture of particle and wave eVs = hf − Es ,
behaviors. At the turn of the twentieth century,
physicists assumed that particle and wave phenom- where Es is the amount of energy required by an
ena were completely distinct. Young had shown that electron to penetrate the surface of the cathode and
light could undergo interference effects such as diffrac- escape.
tion, so it had to be a wave. Since light was a wave
composed of oscillating electric and magnetic fields,
it made sense that when light encountered matter,
it would tend to shake the electrons. It was only
to be expected that something like the photoelectric
effect could happen, with the light shaking the elec-
trons vigorously enough to knock them out of the
atom. The best theoretical estimates, however, were
that light of ordinary intensity would take millions
of years to do the trick — it would take that long
for the electron slowly to absorb enough energy to
escape.
The actual experimental observation of the photo-
electric effect was therefore an embarrassment. It Optical setup.
started up immediately, not after a million years.
Albert Einstein, better known today for the theory
of relativity, was the first to come up with the rad- Setup
ical, and correct, explanation. Einstein simply sug-
gested that in the photoelectric effect, light was be- You can use the Hg gas discharge tube to produce
having as a particle, now called a photon. The beam monochromatic light with the following wavelengths:

144 Lab 45 The Photoelectric Effect

 color wavelength (nm)
ultraviolet 365
violet 405
blue 436
green 546
yellow 578
The diffraction grating splits up the light into these
lines, so you can make one line at a time enter the
photodiode. Slit 1 slides into the slot in the front of
the discharge tube. The lens serves to create focused
images of slit 1 at the photodiode. The lens and
diffraction grating are housed in a single unit, which
is attached to a pair of rods (not shown) projecting
from slit 1. Do not drop the lens and diffraction
grating — I have already damaged one by dropping
it, and they cost $200 to replace. For measurements
with the green and yellow lines, green and yellow
filters are used to help eliminate stray light of other
colors — they stick magnetically on the front of the
collimator tube. Slit 2 and the collimator tube keep
stray light from getting in.
The photodiode module is held on top of a post on
a rotating arm. The ultraviolet line is invisible, but
the front of slit 2 is coated with a material that flu-
oresces in UV light, so you can see where the line
is.
Circuit.
Circuit
The circuit in fig. (a) above is the one shown in
textbooks for this type of experiment. Light comes
in and knocks electrons out of the curved cathode.
If the voltage is turned off, there is no electric field,
so the electrons travel in straight lines; some will
hit the anode, creating a current referred to as the
photocurrent. If the voltage is turned on, the electric
field repels the electrons from the wire electrode, and
the current is reduced or eliminated. The stopping
voltage would be measured by increasing the voltage
until no more current was flowing. We used to use
a setup very similar to this in this course, but it
was difficult to get good data because it was hard to
judge accurately when the current had reached zero.
The circuit we now use, shown in fig. (b), uses a
cute trick to determine the stopping voltage. The
photocurrent transports electrons from the cathode
to the anode, so a net positive charge builds up on
the cathode, and a negative charge on the anode. As
the charge and the voltage increase, the photocur-
rent is reduced, until finally the voltage reaches the
stopping voltage, and no more current can flow. You
then read the voltage off of the voltmeter. When
you have the next color of light shining on the cath-
ode, you momentarily close the switch, discharging
the photodiode, and then take your next measure-
ment. The only disadvantage of this setup is that
you cannot adjust the voltage yourself and see how
the photocurrent varies with voltage.
Setup
Move the housing containing the grating and lens
until you get a good focus at the front of the photo-
diode box. The square side needs to be facing away
from the discharge tube.
Diffraction patterns are supposed to be symmetric,
i.e., the m = 1 and m = −1 maxima should be iden-
tical. In reality, there is something strange about
this setup that can cause the shorter wavelength
lines (especially the UV line) to be extremely dim
on one side. Check which one is brighter on your
apparatus.
Just because the light gets in through slit 2 does
not mean it is getting in to the photodiode. The
original design of the apparatus allowed the photo-
diode module to twist around on its post, and it
had to be adjusted carefully by trial and error. Be-
cause students were getting frustrated with this, I
epoxied the photodiode modules onto their posts in
the right orientation. This makes it impossible to
disassemble the apparatus and put it in its storage
box, but should get rid of the hassles with orient-
ing it. However, you should still check that it’s ori-
ented correctly, because it’s possible that your setup
was a little different from mine, I’m the epoxy can
be cracked by rough handling, and the screw at the
base of the post can also get loose. There are three
145
things you should check to make sure the orientation
is right: (1) Sighting along the tube like a gun, you
should see that it looks like it’s lined up with the
center of the grating. (2) The tube can be lifted out
on a hinge so that you can see the glass photodi-
ode tube inside the box; check that light is actually
falling on the opening on the side of the tube. (3)
Take data using the UV line. If you don’t get a big-
ger voltage for this line than for the others, then the
light is not making it in to the photodiode.
Observations
You can now determine the stopping voltages corre-
sponding to the five different colors of light.
Hints:
The biggest possible source of difficulty is stray
light. The room should be dark when you do
your measurements.
The shortest wavelengths of light (highest fre-
quencies), for which the energy of the pho-
tons is the highest, readily produce photoelec-
trons. The photocurrent is much weaker for
the longer wavelengths. Start with the short-
wavelength line and graduate to the more diffi-
cult, lower frequencies. Don’t forget the filters
for the yellow and green lines!
If the button to zero the voltage doesn’t work,
it is because the batteries are dead.
When you hit the button to zero the voltage, it
may actually pop up high and then come back
down. This is normal. (It’s acting like an RC
circuit with a long RC time constant).
Check the batteries in your photodiode module
before you start, using the two banana plugs
designed for this purpose. If your batteries are
dead, you need to replace them. I’ve also seen
cases where the batteries are on the borderline
at the beginning of the lab, and then die com-
pletely during the lab; in this situation, you’ll
notice that the stopping voltages you’re mea-
suring change over the course of the lab, and
don’t make sense. It won’t hurt to check the
batteries at the end of the lab as well as at the
beginning.
Where the lines hit the white front of slit 2,
they should be sharp, and should not overlap.
You can adjust the focus by moving the lens
and grating in or out. If you can’t get a good
146 Lab 45 The Photoelectric Effect
focus, check and make sure that the square side
of the unit is away from the Hg tube.
The photodiode module can be rotated on its
post so that the light goes straight down the
tube. If you don’t line it up correctly, you’ll
be able to tell because the voltage will creep
up slowly, rather than shooting up to a certain
value and stopping. There is a screw that is
supposed to allow you to lock the photodiode
into position at the correct angle. Make sure
to loosen the screw before trying to aim the
photodiode, and lock it once it’s aimed cor-
rectly. If your photodiode won’t lock in place,
you need to tighten the aluminum post that
forms the base of the box.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
The week before you are to do the lab, briefly famil-
iarize yourself visually with the apparatus.
P1 In the equation eVs = hf − Es , verify that all
three terms have the same units.
P2 Plan how you will analyze your data to deter-
mine Planck’s constant.
Analysis
Extract Planck’s constant from your data, with error
bars (see appendix 4). Is your value consistent with
the accepted value given in your textbook?
Every electron that absorbs a photon acquires a ki-
netic energy equal to hf . Thus it would seem that
if the voltage is less than the stopping voltage, ev-
ery electron should have enough energy to reach the
other electrode. Give two reasons why many elec-
trons do not reach the other electrode even when
the voltage is less than the stopping voltage.
147
46 Electron Diffraction
Apparatus
cathode ray tube (Leybold 555 626)
high-voltage power supply (new Leybold)
100-kΩ resistor with banana-plug connectors
Vernier calipers

Goals
Observe wave interference patterns (diffraction
patterns) of electrons, demonstrating that elec-
trons exhibit wave behavior as well as particle
behavior.

Learn what it is that determines the wave-

length of an electron.

Introduction
The most momentous discovery of 20th-century physics
has been that light and matter are not simply made
of waves or particles — the basic building blocks of
light and matter are strange entities which display
both wave and particle properties at the same time.
In our course, we have already learned about the
experimental evidence from the photoelectric effect The electron diffraction tube. The distance labeled as
showing that light is made of units called photons, 13.5 cm in the figure actually varies from about 12.8 cm
which are both particles and waves. That proba- to 13.8 cm, even for tubes that otherwise appear identical.
bly disturbed you less than it might have, since you
most likely had no preconceived ideas about whether
light was a particle or a wave. In this lab, however,
you will see direct evidence that electrons, which you Method
had been completely convinced were particles, also
What you are working with is basically the same
display the wave-like property of interference. Your
kind of vacuum tube as the picture tube in your tele-
schooling had probably ingrained the particle inter-
vision. As in a TV, electrons are accelerated through
pretation of electrons in you so strongly that you
a voltage and shot in a beam to the front (big end)
used particle concepts without realizing it. When
of the tube, where they hit a phosphorescent coat-
you wrote symbols for chemical ions such as Cl−
ing and produce a glow. You cannot see the electron
and Ca2+ , you understood them to mean a chlorine
beam itself. There is a very thin carbon foil (it looks
atom with one excess electron and a calcium atom
like a tiny piece of soap bubble) near where the neck
with two electrons stripped off. By teaching you to
joins the spherical part of the tube, and the elec-
count electrons, your teachers were luring you into
trons must pass through the foil before crossing over
the assumption that electrons were particles. If this
to the phosphorescent screen.
lab’s evidence for the wave properties of electrons
disturbs you, then you are on your way to a deeper The purpose of the carbon foil is to provide an ultra-
understanding of what an electron really is — both fine diffraction grating — the “grating” consists of
a particle and a wave. the crystal lattice of the carbon atoms themselves!

148 Lab 46 Electron Diffraction

As you will see in this lab, the wavelengths of the
electrons are very short (a fraction of a nanometer),
which makes a conventional ruled diffraction grating
useless — the closest spacing that can be achieved on
a conventional grating is on the order of one microm-
eter. The carbon atoms in graphite are arranged in
sheets, each of which consists of a hexagonal pattern
of atoms like chicken wire. That means they are not
lined up in straight rows, so the diffraction pattern
is slightly different from the pattern produced by a
ruled grating.
Also, the carbon foil consists of many tiny graphite
crystals, each with a random orientation of its crys-
tal lattice. The net result is that you will see a bright
spot surrounded by two faint circles. The two circles
represent cones of electrons that intersect the phos-
phor. Each cone makes an angle θ with respect to
the central axis of the tube, and just as with a ruled
grating, the angle is given by
sin θ = λ/d
where λ is the wavelength of the wave. For a ruled
grating, d would be the spacing between the lines.
In this case, we will have two different cones with
two different θ’s, θ1 and θ2 , corresponding to two
different d0 s, d1 and d2 . Their geometrical meaning
is shown below.
The carbon atoms in the graphite crystal are arranged
hexagonally.
Safety
This lab involves the use of voltages of up to 6000 V.
Do not be afraid of the equipment, however; there
is a fuse in the high-voltage supply that limits the
amount of current that it can produce, so it is not
particularly dangerous. Read the safety checklist on
high voltage in Appendix 8. Before beginning the
lab, make sure you understand the safety rules, ini-
tial them, and show your safety checklist to your
instructor. If you don’t understand something, ask
your instructor for clarification.
In addition to the high-voltage safety precautions,
please observe the following rules to avoid damaging
the apparatus:
The tubes cost $1000. Please treat them with
respect! Don’t drop them! Dropping them would
also be a safety hazard, since they’re vacuum tubes,
so they’ll implode violently if they break.
Do not turn on anything until your instructor
has checked your circuit.
Don’t operate the tube continuously at the
highest voltage values (5000-6000 V). It produces
x-rays when used at these voltages, and the strong
beam also decreases the life of the tube. You can
use the circuit on the right side of the HV supply’s
panel, which limits its own voltage to 5000 V. Don’t
leave the tube’s heater on when you’re not actually
taking data, because it will decrease the life of the
tube.
Setup
You setup will consist of two circuits, a heater circuit
and the high-voltage circuit.
The heater circuit is to heat the cathode, increas-
ing the velocity with which the electrons move in
the metal and making it easier for some of them
to escape from the cathode. This will produce the
friendly and nostalgia-producing yellow glow which
is characteristic of all vacuum-tube equipment. The
heater is simply a thin piece of wire, which acts as
a resistor when a small voltage is placed across it,
producing heat. Connect the heater connections, la-
beled F1 and F2, to the 6-V AC outlet at the back
of the HV supply.
The high-voltage circuit’s job is to accelerate the
electrons up to the desired speed. An electron that
happens to jump out of the cathode will head “down-
hill” to the anode. (The anode is at a higher voltage
than the cathode, which would make it seem like
it would be uphill from the cathode to the anode.
However, electrons have negative charge, so they’re
like negative-mass water that flows uphill.) The high
voltage power supply is actually two different power
supplies in one housing, with a left-hand panel for
one and a right-hand panel for the other. Connect
the anode (A) and cathode (C) to the right-hand
149
panel of the HV supply, and switch the switch on
the HV supply to the right, so it knows you’re using
the right-hand panel.
The following connections are specified in the doc-
umentation, although I don’t entirely understand
what they’re for. First, connect the electrode X to
the same plug as the cathode.1 Also, connect F1 to
C with the wire that has the 100-kΩ resistor spliced
into it. The circuit diagram on page 152 summarizes
all this.
Check your circuit with your instructor before turn-
ing it on!
Observations
You are now ready to see for yourself the evidence of
the wave nature of electrons, observe the diffraction
pattern for various values of the high voltage, and
figure out what determines the wavelength of the
electrons. You will need to do your measurements
in the dark.
You will measure the θ’s, and thus determine the
wavelength, λ, for several different voltages. Each
voltage will produce electrons with a different veloc-
ity, momentum, and energy.
Hints:
While measuring the diffraction pattern, don’t
touch the vacuum tube — the static electric
fields of one’s body seem to be able to perturb
the pattern.
It is easiest to take measurements at the high-
est voltages, where the electrons pack a wallop
and make nice bright rings on the phosphor.
Start with the highest voltages and take data
at lower and lower voltages until you can’t see
the rings well enough to take precise data. To
get unambiguous results, you’ll need to take
data with the widest possible range of voltages.
In order to reach a definite conclusion about
what λ is proportional to, you will need accu-
rate data. Do your best to get good measure-
ments. Pay attention to possible problems in-
curred by viewing the diffraction patterns from
different angles on different occasions. Try re-
peating a measurement more than once, and
seeing how big your random errors are.
1 Ifyou look inside the tube, you can see that X is an extra
electrode sandwiched in between the anode and the cathode.
I think it’s meant to help produce a focused beam.
150 Lab 46 Electron Diffraction
You need to get data down to about 2 or 3
kV in order to get conclusive results from this
experiment. The tubes are not quite identi-
cal, and were not designed to operate at such
low voltages, so they haven’t been tested un-
der those conditions. Experience has shown
that some of the tubes work at lower voltages
than others. The group that has the tube that
works the best at low voltages can share their
low-voltage data with the other groups.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
The week before you are to do the lab, briefly famil-
iarize yourself visually with the apparatus.
Read the safety checklist.
P1 It is not practical to measure θ1 and θ2 directly
with a protractor. Come up with a plan for how to
get the angles indirectly using trigonometry.
The figure shows the vacuum tube as having a par-
ticular shape, which is a sphere with the foil and
phosphor at opposite ends of a diamater. In reality,
the tubes we’re using now are not quite that shape.
To me, they look like they may have been shaped
so that the phosphor surface is a piece of a sphere
centered on the foil. If so, then arc lengths across
the phosphor can be connected to diffraction angles
very simply via the definition of radian measure.
P2 If the voltage difference across which the elec-
trons are accelerated is V , and the known mass and
charge of the electron are m and e, what are the
electrons’ kinetic energy and momentum, in terms
of V , m, and e? (As a numerical check on your re-
sults, you should find that V = 5700 V gives KE =
9.1 × 10−16 J and p = 4.1 × 10−23 kg·m/s.)
P3 All you’re trying to do based on your graphs is
judge which one could be a graph of a proportional-
ity, i.e., a line passing through the origin. Because
of this, you can omit any constant factors from the
equations you found in P1. When you do this, what
do your expressions turn out to be?
P4 Why is it not logically possible for the wave-
length to be proportional to both p and KE? To
both 1/p and 1/KE?
P5 I have suggested plotting λ as a function of
p, KE, 1/p and 1/KE to see if λ is directly propor-
tional to any of them. Once you have your raw data,
how can you immediately rule out two of these four
possibilities and avoid drawing the graphs?
P6 On each graph, you will have two data-points
for each voltage, corresponding to two different mea-
surements of the same wavelength. The two wave-
lengths will be almost the same, but not exactly
the same because of random errors in measuring the
rings. Should you get the wavelengths by combining
the smaller angle with d1 and the larger angle with
d2 , or vice versa?

Analysis
Once you have your data, you can try plotting λ as
a function of, say, the kinetic energy, KE, of the
electrons, and see if it makes something simple like
a straight line. Make sure your graph includes the
origin (see below). You could also try plotting λ
as a function of the electrons’ momentum, p, or as
a function of other quantities such as 1/KE, 1/p,
etc. You can simplify your analysis by leaving out
constant factors.
What does λ seem to be proportional to? Your data
may cover a small enough range of voltage that more
than one graph may look linear. You can rule one
out by checking whether a line fit through the data
points would pass near the origin, as it must for a
proportionality. This is why it is important to have
your graph include the origin.

151
The circuit for the new setup.

152 Lab 46 Electron Diffraction

153
47 The Hydrogen Atom
Apparatus
H gas discharge tube . . . . . . . . . . . . . . . . . . . . . 1/group
Hg gas discharge tube . . . . . . . . . . . . . . . . . . . . 1/group
spectrometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
diffraction grating, 600 lines/mm . . . . . . . . . 1/group
small screwdriver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
black cloth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
piece of plywood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
block of wood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
penlight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1/group
Goals
Observe the visible line spectrum of hydrogen.
Determine the mass of the electron.
Introduction
What’s going on inside an atom? The question would
have seemed nonsensical to physicists before the 20th
century — the word “atom” is Greek for “unsplit-
table,” and there was no evidence for subatomic
particles. Only after Thomson and Rutherford had
demonstrated the existence of electrons and the nu-
cleus did the atom begin to be imagined as a tiny
solar system, with the electrons moving in elliptical
orbits around the nucleus under the influence of its
electric field. The problem was that physicists knew
very well that accelerating charges emit electromag-
netic radiation, as for example in a radio antenna, so
the acceleration of the electrons should have caused
them to emit light, steadily lose energy, and spiral
into the nucleus, all within a microsecond,.
Luckily for us, atoms do not spontaneously shrink
down to nothing, but there was indeed evidence that
atoms could emit light. The spectra emitted by very
hot gases were observed to consist of patterns of dis-
crete lines, each with a specific wavelength. The
process of emitting light always seemed to stop short
of finally annihilating the atom — why? Also, why
were only those specific wavelengths emitted?
In this lab, you will study the spectrum of light emit-
ted by the hydrogen atom, the simplest of all atoms,
with just one proton and one electron. In 1885, be-
fore electrons and protons had even been imagined,
154 Lab 47 The Hydrogen Atom
a Swiss schoolteacher named Johann Balmer discov-
ered that the wavelengths emitted by hydrogen were
related by mysterious ratios of small integers. For in-
stance, the wavelengths of the red line and the blue-
green line form a ratio of exactly 20/27. Balmer even
found a mathematical rule that gave all the wave-
lengths of the hydrogen spectrum (both the visible
ones and the invisible ones that lay in the infrared
and ultraviolet). The formula was completely empir-
ical, with no theoretical basis, but clearly there were
patterns lurking in the seemingly mysterious atomic
spectra.
The first step toward understanding Balmer’s nu-
merology was Einstein’s theory that light consisted
of particles (photons), whose energy was related to
their frequency by the equation Ephoton = hf , or
substituting f = c/λ, Ephoton = hc/λ .
According to this theory, the discrete wavelengths
that had been observed came from photons with spe-
cific energies. It seemed that the atom could exist
only in specific states of specific energies. To get
from an initial state with energy Ei to a final state
with a lower energy Ef , conservation of energy re-
quired the atom to release a photon with an energy
of Ephoton = Ei − Ef .
Not only could the discrete line spectra be explained,
but if the atom possessed a state of least energy
(called a “ground state”), then it would always end
up in that state, and it could not collapse entirely.
Knowing the differences between the energy levels of
the atom, it was not too difficult to figure out the
atomic energy levels themselves. Niels Bohr showed
that they obey a relatively simple equation,
mk 2 e4 1
En = − · 2
2~2 n
where n is an integer labeling the level, k is the
Coulomb constant, e is the fundamental unit of charge,
~ is Planck’s constant over 2π, and me is the mass of
the electron. All the energies of the photons in the
emission spectrum could now be explained as differ-
ences in energy between specific states of the atom.
For instance the four visible wavelengths observed by
Balmer all came from cases where the atom ended up
in the n = 2 state, dropping down from the n = 3,
4, 5, and 6 states.
Although the equation’s sheer size may appear for-
midable, keep in mind that the quantity in paren-
theses is just a numerical constant, and the variation
of energy from one level to the next is of the very
simple mathematical form 1/n2 . It was because of
this basic simplicity that the wavelength ratios like
20/27 occurred. The minus sign occurs because the
equation includes both the electron’s potential en-
ergy and its kinetic energy, and the standard choice
of a reference-level for the potential energy results
in negative values.
Along with the nice formulas came a whole new
set of subversive concepts: that nature is random
in certain ways, that the building blocks of nature
are both particles and waves, and that subatomic
particles do not follow well-defined trajectories as
they travel through space. Today these ridiculous-
sounding ideas are taken for granted by working physi-
cists, and we are so sure of the theory behind Bohr’s
equation that it is now used as one of the most accu-
rate ways of determining the mass of the electron. In
a previous lab, you measured the charge-to-mass ra-
tio of the electron, but like the experiment by Thom-
son on which it was based, that technique was un-
able to give the charge and mass separately. Mod-
ern techniques allow us to measure wavelengths of
light, and therefore energies of photons, with high
precision, so if all the other fundamental constants
in Bohr’s equation are known, we can solve for the
mass of the electron. This lab is really the only ex-
ample of a high-precision experiment that you’ll do
in this course — done correctly, it allows the de-
termination of the electron’s mass to five significant
figures!
Method
The apparatus you will use to observe the spectrum
of hydrogen is shown in the figure. For a given wave-
length, the grating produces diffracted light at many
different angles: a central zeroth-order line at θ = 0,
first-order lines on both the left and right, and so on
through higher-order lines at larger angles. The line
of order m occurs at an angle satisfying the equation
mλ = d sin θ.
To measure a wavelength, you will move the tele-
scope until the diffracted first-order image of the slit
is lined up with the telescope’s cross-hairs and then
read off the angle. Note that the angular scale on
the table of the spectroscope actually gives the angle
labeled α in the figure, not θ.
Sources of systematic errors
There are three sources of systematic error:
angular scale out of alignment: If the angular
scale is out of alignment, then all the angles
will be off by a constant amount.
factory’s calibration of d: The factory that
made the grating labeled it with a certain spac-
ing (in lines per millimeter) which can be con-
verted to d (center-to-center distance between
lines). But their manufacturing process is not
all that accurate, so the actual spacing of the
lines is a little different from what the label
says.
orientation of the grating: Errors will be caused
if the grating is not perpendicular to the beam
from the collimator, or if the lines on the grat-
ing are not vertical (perpendicular to the plane
of the circle).
Eliminating systematic errors
A trick to eliminate the error due to misalignment of
the angular scale is to observe the same line on both
the right and the left, and take θ to be half the differ-
ence between the two angles, i.e., θ = (αR − αL )/2.
Because you are subtracting two angles, any source
of error that adds a constant offset onto the angles
is eliminated. A few of the spectrometers have their
angular scales out of alignment with the collimators
by as much as a full degree, but that’s of absolutely
no consequence if this technique is used.
Regarding the calibration of d, the first person who
ever did this type of experiment simply had to make
a diffraction grating whose d was very precisely con-
structed. But once someone has accurately mea-
sured at least one wavelength of one emission line
of one element, one can simply simply to determine
the spacing, d, of any grating using a line whose
wavelength is known.
You might think that these two tricks would be enough
to get rid of any error due to misorientation of the
grating, but they’re not. They will get rid of any
error of the form θ → θ + c or sin θ → c sin θ, but
misorientation of the grating produces errors of the
form sin θ → sin θ + c. The adjustment of the orien-
tation of the grating is described later.
Optics
The figure below shows the optics from the side, with
the telescope simply looking down the throat of the
collimator at θ = 0. You are actually using the op-
155
tics to let you see an image of the slit, not the tube
itself. The point of using a telescope is that it pro-
vides angular magnification, so that a small change
in angle can be seen visually.
A lens is used inside the collimator to make the light
from the slit into a parallel beam. This is important,
because we are using mλ = d sin θ to determine the
wavelength, but this equation was derived under the
assumption that the light was coming in as a parallel
beam. To make a parallel beam, the slit must be
located accurately at the focal point of the lens. This
adjustment should have already been done, but you
will check later and make sure. A further advantage
of using a lens in the collimator is that a telescope
only works for objects far away, not nearby objects
from which the reflected light is diverging strongly.
The lens in the collimator forms a virtual image at
infinity, on which the telescope can work.
The objective lens of the telescope focuses the light,
forming a real image inside the tube. The eyepiece
then acts like a magnifying glass to let you see the
image. In order to see the cross-hairs and the image
of the slit both in focus at the same time, the cross-
hairs must be located accurately at the focal point
of the objective, right on top of the image.
Setup
Turn on both gas discharge tubes right away, to let
them get warmed up.
Adjusting the optics at the start of the experiment is
vital. You do not want to fail to get the adjustments
right and then spend several frustrating hours trying
fruitlessly to make your observations.
First you must check that the cross-hairs are at the
focal point of the objective. If they are, then the im-
age of the slits formed by the objective will be at the
same point in space as the crosshairs. You’ll be able
to focus your eye on both simultaneously, and there
will be no parallax error depending on the exact po-
sition of your eye. The easiest way to check this is
to look through the telescope at something far away
(& 50 m), and move your head left and right to see if
the crosshairs move relative to the image. Slide the
eyepiece in and out to achieve a comfortable focus.
If this adjustment is not correct, you may need to
move the crosshairs in or out; this is done by sliding
the tube that is just outside the eyepiece tube. (You
need to use the small screwdriver to loosen the screw
on the side, which is recessed inside a hole. The hole
may have a dime-sized cover over it.)
156 Lab 47 The Hydrogen Atom
The white plastic pedestal should have already been
adjusted correctly to get the diffraction grating ori-
ented correctly in three dimensions, but you should
check it carefully. There are some clever features
built into the apparatus to help in accomplishing
this. As shown in the figure, there are three axes
about which the grating could be rotated. Rotation
about axis 1 is like opening a door, and this is ac-
complished by rotating the entire pedestal like a lazy
Susan. Rotation about axes 2 (like folding down a
tailgate) and 3 are accomplished using the tripod of
screws underneath the pedestal. The eyepiece of the
telescope is of a type called a Gauss eyepiece, with
a diagonal piece of glass in it. When the grating is
oriented correctly about axes 1 and 2 and the tele-
scope is at θ = 0, a beam of light that enters through
the side of the eyepiece is partially reflected to the
grating, and then reflected from the grating back to
the eye. If these two axes are correctly adjusted, the
reflected image of the crosshairs is superimposed on
the crosshairs.
First get a rough initial adjustment of the pedestal
by moving the telescope to 90 degrees and sight-
ing along it like a gun to line up the grating. Now
loosen the screw (not shown in the diagram) that
frees the rotation of the pedestal. Put a desk lamp
behind the slits of the collimator, line up the tele-
scope with the m = 0 image (which may not be
exactly at α = 180 degrees), remove the desk lamp,
cover the whole apparatus with the black cloth, and
position a penlight so that it shines in through the
hole in the side of the eyepiece. Adjust axes 1 and
2. If you’re far out of adjustment, you may see part
of a circle of light, which is the reflection of the pen-
light; start by bringing the circle of light into your
field of view. When you’re done, tighten the screw
that keeps the pedestal from rotating. The pedestal
is locked down to the tripod screws by the tension
in a spring, which keeps the tips of two of the screws
secure in dimples underneath the platform. Don’t
lower the screws too much, or the pedestal will no
longer stay locked; make a habit of gently wiggling
the pedestal after each adjustment to make sure it’s
not floating loose. Two of the spectrometers have
the diagonal missing from their eyepieces, so if you
have one of those, you’ll have to borrow an eyepiece
from another group to do this adjustment.
For the adjustment of axis 3, place a piece of masking
tape so that it covers exactly half of the slits of the
collimator. Put the Hg discharge tube behind the
slits. The crosshairs should be near the edge of the
tape in the m = 0 image. Move the telescope out to a
large angle where you see one of the high-m Hg lines,
and adjust the tripod screws so that the crosshairs
are at the same height relative to the edge of the
tape.
Observations
Now put the Hg tube behind the collimator. Make
sure the hottest part of the tube is directly in front of
the slits; you will need to use the piece of plywood to
raise the spectrometer to the right height. You want
the tube as close to the slits as possible, and lined
up with the slits as well as possible; you can adjust
this while looking through the telescope at an m = 1
line, so as to make the line as bright as possible.
If your optics are adjusted correctly, you should be
able to see the microscopic bumps and scratches on
the knife edges of the collimator, and there should
be no parallax of the crosshairs relative to the image
of the slits.
Here is a list of the wavelengths of the visible Hg
lines, in nm, to high precision:
404.656 violet There is a dimmer violet
line nearby at 407.781 nm.
435.833 blue
491.604 blue-green dim
546.074 green There is also a dimmer
blue-green line.
yellow This is actually a complex
set of lines, so it’s not use-
ful for calibration.
You’ve seen the mercury spectrum before, in the
photoelectric effect lab, but you will notice some dif-
ferences here. You will be able to see various dim-
mer lines as well as the bright ones, and you want
to avoid mistaking those for the bright ones that
we’re using for calibration. Also, as noted above,
some lines that appear to be single lines in a low-
resolution spectrum are actually mutliple lines. The
table gives the wavelengths in vacuum. Although
we’re doing the lab in air, our goal is to find what
the hydrogen wavelengths would have been in vac-
uum; by calibrating using vacuum wavelengths for
mercury, we end up getting vacuum wavelengths for
hydrogen as well.
Start by making sure that you can find all of the
mercury lines in the correct sequence — if not, then
you have probably found some first-order lines and
some second-order ones. If you can find some lines
but not others, use your head and search for them
in the right area based on where you found the lines
you did see. You may see various dim, fuzzy lights
through the telescope — don’t waste time chasing
these, which could be coming from other tubes or
from reflections. The real lines will be bright, clear
and well-defined. By draping the black cloth over
the discharge tube and the collimator, you can get
rid of stray light that could cause problems for you
or others. Put a box behind the discharge tube to
block the light coming out through the hole in the
back as well. Do a quick and rough check that the
angles at which you observe the violet line are close
to the approximate one predicted in prelab question
P1.
Now try swapping in the hydrogen tube in place of
the mercury tube, and go through a similar process
of acquainting yourself with the four lines in its vis-
ible spectrum, which are as follows:
violet dim
purple
blue-green
red
Again you’ll again have to make sure the hottest
part of the tube is in front of the collimator; this
requires putting books and/or blocks of wood under
the discharge tube.
By coincidence, three of the mercury lines lie very
close to three of the hydrogen lines: violet next to
violet, blue next to purple, and blue-green next to
blue-green. We’ll exploit this coincidence to reduce
systematic errors. The idea is that if we have a hy-
drogen wavelength, λ, that we want to measure, and
a nearby known line λc from the mercury spectrum
that we can use for calibration, then we can sidestep
the determination of the grating’s spacing entirely
by using the relation
sin θ
λ= λc .
sin θc
Although we’ve tried hard to eliminate systematic
errors through careful adjustment of the optics, some
errors will still remain. But these errors vary smooth-
ly with angle, so by calibrating against known lines
that occur at nearly the same angle, we can mini-
mize their effects. For each line, you’ll need a total
of four angles: αL and αR for the calibration line
at m = ±1, and a similar pair of angles for the hy-
drogen line. Note that the knob that adjusts the
width of the slit only moves the right-hand knife-
edge, which appears to be on the left in the inverted
image. Therefore, adjusting the slit changes the lo-
cation of the center of the slit, and such adjustments
should not be done between the αL and αR measure-
ments of the same line. My experience is that the
dimness of the violet hydrogen line makes its wave-
157
length difficult to measure accurately, so including
the violet-violet pair actually worsens the quality of
the final result. Therefore you end up just measuring
two pairs (eight angles).
The angles are measured using a vernier scale, which
is similar to the one on the vernier calipers you have
already used in the first-semester lab course. Your
final reading for an angle will consist of degrees plus
minutes. (One minute of arc, abbreviated 1’, is 1/60
of a degree.) The main scale is marked every 30
minutes. Your initial, rough reading is obtained by
noting where the zero of the vernier scale falls on the
main scale, and is of the form “xxx ◦ 0’ plus a little
more” or “xxx ◦ 30’ plus a little more.” Next, you
should note which line on the vernier scale lines up
most closely with one of the lines on the main scale.
The corresponding number on the vernier scale tells
you how many minutes of arc to add for the “plus a
little more.”
As a check on your results, everybody in your group
should take independent readings of every angle you
measure in the lab, nudging the telescope to the side
after each reading. Once you have independent re-
sults for a particular angle, compare them. If they’re
consistent to within one or two minutes of arc, aver-
age them. If they’re not consistent, figure out what
went wrong.
Prelab
The point of the prelab questions is to make sure
you understand what you’re doing, why you’re do-
ing it, and how to avoid some common mistakes. If
you don’t know the answers, make sure to come to
my office hours before lab and get help! Otherwise
you’re just setting yourself up for failure in lab.
The week before you are to do the lab, briefly famil-
iarize yourself visually with the apparatus.
P1 The nominal (and not very accurate) spacing
of the grating is stated as 600 lines per millimeter.
From this information, find d, and predict the angles
αL and αR at which you will observe the 404.656 nm
violet mercury line.
P2 Make sure you understand the first three vernier
readings in the figure, and then interpret the fourth
one.
P3 In what sequence do you expect to see the Hg
lines on each side? Make a drawing showing the
sequence of the angles as you go out from θ=0.
P4 The visible lines of hydrogen come from the
158 Lab 47 The Hydrogen Atom
3 → 2, 4 → 2, 5 → 2, and 6 → 2 transitions. Based
on E = hf , which of these should correspond to
which colors?
P5 Based on the Bohr equation, predict the ratio
λblue−green
λpurple
for hydrogen, expressing your answer as the ratio of
two integers.
Self-Check
Before leaving lab, make sure that your wavelengths
are consistent with your prediction from prelab ques-
tion P5, to a precision of no worse than about one
part per thousand.
Analysis
Throughout your analysis, remember that this is
a high-precision experiment, so you don’t want to
round off to less than five significant figures.
We assume that the following constants are already
known:
e = 1.6022 × 10−19 C
k = 8.9876 × 109 N·m2 /C2
h = 6.6261 × 10−34 J·s
c = 2.9979 × 108 m/s
The energies of the four types of visible photons
emitted by a hydrogen atom equal En − E2 , where
n = 3, 4, 5, and 6. Using the Bohr equation, we have
 
1 1
Ephoton = A − 2 ,
4 n
where A is the expression from the Bohr equation
that depends on the mass of the electron. From the
two lines you’ve measured, extract a value for A.
If your data passed the self-check above, then you
should find that these values for A agree to no worse
than a few parts per thousand at worst. Compute
an average value of A, and extract the mass of the
electron, with error bars.
Finally, there is a small correction that should be
made to the result for the mass of the electron be-
cause actually the proton isn’t infinitely massive com-
pared to the electron; in terms of the quantity m
given by the equation on page 154, the mass of the
electron, me , would actually be given by me = m/(1−
m/mp ), where mp is the mass of the proton, 1.6726×
10−27 kg.
The spectrometer

Optics.

Orienting the grating.

Prelab question 2.

159
48 The Michelson Interferometer
Apparatus
Michelson interferometer . . . . . . . . . . . . . . . . . .1/group
Na and H gas discharge tubes . . . . . . . . . . . . 1/group
tools inside drawer . . . . . . . . . . . . . . . . . . . . 1 set/group
2 × 4 piece of wood . . . . . . . . . . . . . . . . . . . . . . . 3/group
colored filters (Cambosco and others)
Goals
Determine the wavelength of a line of the emis-
sion spectrum of sodium or hydrogen.
The Michelson interferometer is a device for measur-
ing the wavelength of light, used most famously in
the Michelson-Morley experiment of 1887, which was
later interpreted as disproving the existence of the
luminiferous aether and supporting Einstein’s theory
of special relativity.
As shown in the figure, the idea is to take a beam
of light from the source, split it into two perpendic-
ular beams, send it to two mirrors, and then recom-
bine the beams again. If the two light waves are in
phase when recombined, they will reinforce, but if
they are out of phase, they will cancel. Since the
two waves originated from the splitting of a single
wave, the only reason they would be out of phase
was if the lengths of the two arms of the apparatus
were unequal. Mirror A is movable, and the distance
through which it moves can be controlled and mea-
sured extremely accurately using a micrometer con-
nected to the mirror via a lever. If mirror A is moved
160 Lab 48 The Michelson Interferometer
by distance equal to a quarter of a wavelength of the
light, the total round-trip distance traveled by the
wave is changed by half a wavelength, which switches
from constructive to destructive interference, or vice
versa. Thus if the mirror is moved by a distance d,
and you see the light go through n complete cycles
of appearance and disappearance, you can conclude
that the wavelength of the light was λ = 2d/n.
To make small and accurate adjustments of the mir-
ror easier to do, the micrometer is connected to it
through a level that reduces the amount of move-
ment by a factor k, approximately equal to 5.23;
the micrometer reads the bigger distance D = kd
that it actually travels itself, so the wavelength is
λ = 2D/kn.
Another trick to make the apparatus easier to use
is that the mirrors A and B are slightly curved.
This means that instead of seeing a field of light
that varies uniformly between dark and bright as
you turn the knob, instead you see a set of concen-
tric rings (called fringes), which expand or contract
depending on which direction you turn the knob.
Turn on the sodium discharge tube, and let it warm
up until it’s yellow.
Remove the drawer from the box, and take out the
tool kit. Unscrew the screws on the bottom of the
box that lock the interferometer to the floor of the
box, and very carefully take the instrument out of
the box. Screw the two aluminum legs into the bot-
tom of the interferometer, and lay a piece of wood
flat under the third leg, which is a threaded rod; this
makes the apparatus level.
Place the discharge tube near the entrance window
of the apparatus. If you look through the viewing
window, you will see the image of the tube itself,
reflected through the mirrors. To make this into a
uniform circle of light, place the ground glass screen
(inside the bag of tools) in the bracket at the en-
trance window.
Mirror B needs to be perfectly perpendicular to mir-
ror A, and its vertical plane needs to be matched to
mirror A’s. This is adjusted using the knobs on mir-
ror B, one for vertical adjustment and one for hor-
izontal. A rough initial adjustment can be done by
aligning the two images of the circular entrance win-
dow. You can then hang the metal pointer (from the
bag of tools) on the top of the ground glass screen,
and do a better adjustment so that the two images of make more progress in measuring lines of the hydro-
the pointer’s tip coincide. You should now see a set gen spectrum accurately.
of very fine concentric circular interference fringes,
centered on a point outside of the field of view. The
final, fine adjustment is obtained by bringing the
center of this pattern to the center of the field of
view.
The micrometer has a millimeter scale running from
0 to 25 mm, with half-millimeter divisions on the
bottom. To take a reading on it, first read the
number of millimeters and half-millimeters based on
where the edge of the cylindrical rotating part lies on
this scale. Then add on the reading from the vernier
scale that runs around the circumference of the ro-
tating part, which runs from 0.000 to 0.500 mm. You
should be able to estimate to the nearest thousandth
of a millimeter (tenth of a vernier division).
While looking at the interference fringes, turn the
knob on the micrometer. You will see them either
expand like smoke rings, or contract and disappear
into the center, depending on which way you turn
the knob. Rotate the knob while counting about 50
to 100 fringes, and record the two micrometer read-
ings before and after. The difference between these
is D. It helps if you prop your head on the table,
and move the micrometer knob smoothly and contin-
uously. Moving your head disturbs the pattern, and
halting the micrometer knob tends to cause backlash
that confuses the count of fringes by plus or minus
one.
It has been an ongoing project to get these spec-
trometers back in operation and fully calibrated for
the first time in many years. In spring 2006, my stu-
dents in physics 223 gave them a thorough test drive.
In spring 2007, we started taking data to determine
k accurately for each spectrometer, using the known
wavelength of the sodium emission line at 589 nm.
Their data are on sheets inside each spectrometer’s
box. That class also experimented with using the
apparatus to measure the wavelengths of some lines
in the spectrum of hydrogen, which is of some fun-
damental interest because it is the simplest of all
atoms. Since hydrogen’s spectrum, unlike sodium’s,
includes several different visible lines of similar in-
tensity, this required using colored filters to select
the desired line. They found that filter #2 from
the Cambosco box worked well for the red line, and
#8 for the blue-green line. Lines with short wave-
lengths were more difficult to do. For the next class
that does the lab, my goal is to accumulate more
calibration data, so we can start to detect whether
certain data points are off because of ±1 errors in
counting the number of fringes. I would also like to

161
 Appendix 1: Format of Lab Writeups
Lab reports must be three pages or less, not counting
your raw data. The format should be as follows:
Title
Raw data — Keep actual observations separate from
what you later did with them.
These are the results of the measurements you take
down during the lab, hence they come first. You
should clearly mark the beginning and end of your
raw data, so I don’t have to sort through many pages
to find your actual presentation of your work, below.
Write your raw data directly in your lab book; don’t
write them on scratch paper and recopy them later.
Don’t use pencil. The point is to separate facts from
opinions, observations from inferences.
Procedure — Did you have to create your own
methods for getting some of the raw data?
Do not copy down the procedure from the manual.
In this section, you only need to explain any meth-
ods you had to come up with on your own, or cases
where the methods suggested in the handout didn’t
work and you had to do something different. Do not
discuss how you did your calculations here, just how
you got your raw data.
Abstract — What did you find out? Why is it im-
portant?
The “abstract” of a scientific paper is a short para-
graph at the top that summarizes the experiment’s
results in a few sentences. If your results deviated
from the ideal equations, don’t be afraid to say so.
After all, this is real life, and many of the equa-
tions we learn are only approximations, or are only
valid in certain circumstances. However, (1) if you
simply mess up, it is your responsibility to realize
it in lab and do it again, right; (2) you will never
get exact agreement with theory, because measure-
ments are not perfectly exact — the important issue
is whether your results agree with theory to roughly
within the error bars.
The abstract comes first in your writeup, but you’ll
write it last, so leave a little space for it.
The abstract is not a statement of what you hoped
to find out. It’s a statement of what you did find
out. It’s like the brief statement at the beginning
of a debate: “The U.S. should have free trade with
China.” It’s not this: “In this debate, we will discuss
whether the U.S. should have free trade with China.”
162 Lab Appendix 1: Format of Lab Writeups
If this is a lab that has just one important numerical
result (or maybe two or three of them), put them
in your abstract, with error bars where appropriate.
There should normally be no more than two to four
numbers here. Do not recapitulate your raw data
here — this is for your final results.
If you’re presenting a final result with error bars,
make sure that the number of significant figures is
consistent with your error bars. For example, if you
write a result as 323.54 ± 6 m/s, that’s wrong. Your
error bars say that you could be off by 6 in the ones’
place, so the 5 in the tenths’ place and the four in
the hundredths’ place are completely meaningless.
If you’re presenting a number in scientific notation,
with error bars, don’t do it like this
1.234 × 10−89 m/s ± 3 × 10−92 m/s ,
do it like this
(1.234 ± 0.003) × 10−89 m/s ,
so that we can see easily which digit of the result the
error bars apply to.
Justification and Reasoning — Convince me of
what you claimed in your abstract.
Cconvince me that the statements you made about
your results in the abstract follow logically from your
data. This will typically involve both calculations
and logical arguments. Continuing the debate meta-
phor, if your abstract said the U.S. should have free
trade with China, this is the rest of the debate, where
you convince me, based on data and logic, that we
should have free trade.
In your calculations, the more clearly you show what
you did, the easier it is for me to give you partial
credit if there is something wrong with your final re-
sult. If you have a long series of similar calculations,
you may just show one as a sample. If your prelab
involved deriving equations that you will need, re-
peat them here without the derivation. Try to lay
out complicated calculations in a logical way, go-
ing straight down the page and using indentation to
make it easy to understand. When doing algebra,
try to keep everything in symbolic form until the
very end, when you will plug in numbers.
Model Lab Writeup
Comparison of Heavy and Light Falling Objects
- Galileo Galilei
Raw Data
(Galileo’s original, somewhat messy notes go here.)
He does not recopy the raw data to make them look
nicer, or mix calculations with raw data.
the cannon ball leading at the bottom by a slight
margin (1 hand’s breadth), but we could not be sure.
It is possible that the musket ball was just notice-
ably affected by air resistance. In any case, the Aris-
totelian theory is clearly wrong, since it predicts that
the cannon ball, which was 400 times heavier, would
have taken one 400th the time to hit the ground.

Procedure
We followed the procedure in the lab manual with
the following additions: (1) To make sure both ob-
jects fell at the same time, we put them side by side
on a board and then tipped the board. (2) We waited
until there was no wind.

Abstract
We dropped a cannon ball weighing two hundred
pounds and a musket ball weighing half a pound si-
multaneously from the same height. Both hit the
ground at nearly the same time. This contradicts
Aristotle’s theory that heavy objects always fall faster
than light ones.

height of drop = 200 ± 4 cubits

amount by which cannon
ball was ahead at the bottom < 1 hand’s breadth

Justification and Reasoning

From a point 100 cubits away from the base of the
tower, the top was at a 63 ◦ angle above horizontal.
The height of the tower was therefore

100 cubits × tan 63 ◦ = 200 cubits.

We estimated the accuracy of the 100-cubit horizon-
tal measurement to be ±2 cubits, with random errors
mainly from the potholes in the street, which made
it difficult to lay the cubit-stick flat. If it was 102
cubits instead of 100, our result for the height of the
tower would have been 204 cubits, so our error bars
on the height are ±4 cubits.
It is common knowledge that a feather falls more
slowly than a stone, but our experiment shows that
heavy objects do not always fall much more rapidly.
We do not have any data on feathers, but we sug-
gest that extremely light objects like feathers are
strongly affected by air resistance, which would be
nearly negligible for a cannonball. We think we saw

163
 Appendix 2: Basic Error Analysis
No measurement is perfectly ex-
act.
One of the most common misconceptions about sci-
ence is that science is “exact.” It is always a strug-
gle to get beginning science students to believe that
no measurement is perfectly correct. They tend to
think that if a measurement is a little off from the
“true” result, it must be because of a mistake — if
a pro had done it, it would have been right on the
mark. Not true!
What scientists can do is to estimate just how far
off they might be. This type of estimate is called
an error bar, and is expressed with the ± symbol,
read “plus or minus.” For instance, if I measure my
dog’s weight to be 52 ± 2 pounds, I am saying that
my best estimate of the weight is 52 pounds, and I
think I could be off by roughly 2 pounds either way.
The term “error bar” comes from the conventional
way of representing this range of uncertainty of a
measurement on a graph, but the term is also used
when no graph is involved.
Some very good scientific work results in measure-
ments that nevertheless have large error bars. For
instance, the best measurement of the age of the uni-
verse is now 15 ± 5 billion years. That may not seem
like wonderful precision, but the people who did the
measurement knew what they were doing. It’s just
that the only available techniques for determining
the age of the universe are inherently poor.
Even when the techniques for measurement are very
precise, there are still error bars. For instance, elec-
trons act like little magnets, and the strength of a
very weak magnet such as an individual electron is
customarily measured in units called Bohr magne-
tons. Even though the magnetic strength of an elec-
tron is one of the most precisely measured quantities
ever, the best experimental value still has error bars:
1.0011596524 ± 0.0000000002 Bohr magnetons.
There are several reasons why it is important in sci-
entific work to come up with a numerical estimate
of your error bars. If the point of your experiment
is to test whether the result comes out as predicted
by a theory, you know there will always be some
disagreement, even if the theory is absolutely right.
You need to know whether the measurement is rea-
sonably consistent with the theory, or whether the
discrepancy is too great to be explained by the lim-
164 Lab Appendix 2: Basic Error Analysis
itations of the measuring devices.
Another important reason for stating results with er-
ror bars is that other people may use your measure-
ment for purposes you could not have anticipated.
If they are to use your result intelligently, they need
to have some idea of how accurate it was.
Error bars are not absolute limits.
Error bars are not absolute limits. The true value
may lie outside the error bars. If I got a better scale I
might find that the dog’s weight is 51.3±0.1 pounds,
inside my original error bars, but it’s also possible
that the better result would be 48.7 ± 0.1 pounds.
Since there’s always some chance of being off by a
somewhat more than your error bars, or even a lot
more than your error bars, there is no point in be-
ing extremely conservative in an effort to make ab-
solutely sure the true value lies within your stated
range. When a scientist states a measurement with
error bars, she is not saying “If the true value is
outside this range, I deserve to be drummed out of
the profession.” If that was the case, then every sci-
entist would give ridiculously inflated error bars to
avoid having her career ended by one fluke out of
hundreds of published results. What scientists are
communicating to each other with error bars is a
typical amount by which they might be off, not an
upper limit.
The important thing is therefore to define error bars
in a standard way, so that different people’s state-
ments can be compared on the same footing. By
convention, it is usually assumed that people esti-
mate their error bars so that about two times out of
three, their range will include the true value (or the
results of a later, more accurate measurement with
an improved technique).
Random and systematic errors.
Suppose you measure the length of a sofa with a
tape measure as well as you can, reading it off to
the nearest millimeter. If you repeat the measure-
ment again, you will get a different answer. (This
is assuming that you don’t allow yourself to be psy-
chologically biased to repeat your previous answer,
and that 1 mm is about the limit of how well you
can see.) If you kept on repeating the measurement,
you might get a list of values that looked like this:
203.1 cm 203.4 202.8 203.3 203.2
203.4 203.1 202.9 202.9 203.1
Variations of this type are called random errors, be-
cause the result is different every time you do the
measurement.
The effects of random errors can be minimized by av-
eraging together many measurements. Some of the
measurements included in the average are too high,
and some are too low, so the average tends to be
better than any individual measurement. The more
measurements you average in, the more precise the
average is. The average of the above measurements
is 203.1 cm. Averaging together many measurements
cannot completely eliminate the random errors, but
it can reduce them.
On the other hand, what if the tape measure was a
little bit stretched out, so that your measurements
always tended to come out too low by 0.3 cm? That
would be an example of a systematic error. Since
the systematic error is the same every time, aver-
aging didn’t help us to get rid of it. You probably
had no easy way of finding out exactly the amount
of stretching, so you just had to suspect that there
might a systematic error due to stretching of the
tape measure.
Some scientific writers make a distinction between
the terms “accuracy” and “precision.” A precise
measurement is one with small random errors, while
an accurate measurement is one that is actually close
to the true result, having both small random errors
and small systematic errors. Personally, I find the
distinction is made more clearly with the more mem-
orable terms “random error” and “systematic error.”
The ± sign used with error bars normally implies
that random errors are being referred to, since ran-
dom errors could be either positive or negative, whereas
miles by gallons, to get your final result. When you
systematic errors would always be in the same direc-
communicate your result to someone else, they are
tion.
completely uninterested in how accurately you mea-
sured the number of miles and how accurately you
measured the gallons. They simply want to know
The goal of error analysis how accurate your final result was. Was it 22 ± 2
Very seldom does the final result of an experiment mi/gal, or 22.137 ± 0.002 mi/gal?
come directly off of a clock, ruler, gauge or meter.
Of course the accuracy of the final result is ulti-
It is much more common to have raw data consist-
mately based on and limited by the accuracy of your
ing of direct measurements, and then calculations
raw data. If you are off by 0.2 gallons in your mea-
based on the raw data that lead to a final result.
surement of the amount of gasoline, then that amount
As an example, if you want to measure your car’s
of error will have an effect on your final result. We
gas mileage, your raw data would be the number of
say that the errors in the raw data “propagate” through
gallons of gas consumed and the number of miles
the calculations. When you are requested to do “er-
you went. You would then do a calculation, dividing
ror analysis” in a lab writeup, that means that you

165
are to use the techniques explained below to deter-
mine the error bars on your final result. There are
two sets of techniques you’ll need to learn:
techniques for finding the accuracy of your raw
data
techniques for using the error bars on your raw
data to infer error bars on your final result
Estimating random errors in raw
data
We now examine three possible techniques for es-
timating random errors in your original measure-
ments, illustrating them with the measurement of
the length of the sofa.
Method #1: Guess
If you’re measuring the length of the sofa with a
metric tape measure, then you can probably make a
reasonable guess as to the precision of your measure-
ments. Since the smallest division on the tape mea-
sure is one millimeter, and one millimeter is also near
the limit of your ability to see, you know you won’t
be doing better than ± 1 mm, or 0.1 cm. Making al-
lowances for errors in getting tape measure straight
and so on, we might estimate our random errors to
be a couple of millimeters.
Guessing is fine sometimes, but there are at least two
ways that it can get you in trouble. One is that stu-
dents sometimes have too much faith in a measuring
device just because it looks fancy. They think that
a digital balance must be perfectly accurate, since
unlike a low-tech balance with sliding weights on it,
it comes up with its result without any involvement
by the user. That is incorrect. No measurement is
perfectly accurate, and if the digital balance only
displays an answer that goes down to tenths of a
gram, then there is no way the random errors are
any smaller than about a tenth of a gram.
Another way to mess up is to try to guess the error
bars on a piece of raw data when you really don’t
have enough information to make an intelligent esti-
mate. For instance, if you are measuring the range
of a rifle, you might shoot it and measure how far
the bullet went to the nearest centimeter, conclud-
ing that your random errors were only ±1 cm. In
reality, however, its range might vary randomly by
fifty meters, depending on all kinds of random fac-
tors you don’t know about. In this type of situation,
you’re better off using some other method of esti-
mating your random errors.
166 Lab Appendix 2: Basic Error Analysis
Method #2: Repeated Measurements and the Two-
Thirds Rule
If you take repeated measurements of the same thing,
then the amount of variation among the numbers can
tell you how big the random errors were. This ap-
proach has an advantage over guessing your random
errors, since it automatically takes into account all
the sources of random error, even ones you didn’t
know were present.
Roughly speaking, the measurements of the length
of the sofa were mostly within a few mm of the av-
erage, so that’s about how big the random errors
were. But let’s make sure we are stating our error
bars according to the convention that the true result
will fall within our range of errors about two times
out of three. Of course we don’t know the “true”
result, but if we sort out our list of measurements
in order, we can get a pretty reasonable estimate of
our error bars by taking half the range covered by
the middle two thirds of the list. Sorting out our list
of ten measurements of the sofa, we have
202.8 cm 202.9 202.9 203.1 203.1
203.1 203.2 203.3 203.4 203.4
Two thirds of ten is about 6, and the range covered
by the middle six measurements is 203.3 cm - 202.9
cm, or 0.4 cm. Half that is 0.2 cm, so we’d esti-
mate our error bars as ±0.2 cm. The average of the
measurements is 203.1 cm, so your result would be
stated as 203.1 ± 0.2 cm.
One common mistake when estimating random er-
rors by repeated measurements is to round off all
your measurements so that they all come out the
same, and then conclude that the error bars were
zero. For instance, if we’d done some overenthu-
siastic rounding of our measurements on the sofa,
rounding them all off to the nearest cm, every single
number on the list would have been 203 cm. That
wouldn’t mean that our random errors were zero!
The same can happen with digital instruments that
automatically round off for you. A digital balance
might give results rounded off to the nearest tenth of
a gram, and you may find that by putting the same
object on the balance again and again, you always
get the same answer. That doesn’t mean it’s per-
fectly precise. Its precision is no better than about
±0.1 g.
Method #3: Repeated Measurements and the Stan-
dard Deviation
The most widely accepted method for measuring er-
ror bars is called the standard deviation. Here’s how
the method works, using the sofa example again.
(1) Take the average of the measurements.
average = 203.1 cm
(2) Find the difference, or “deviation,” of each mea-
surement from the average.
−0.3 cm −0.2 −0.2 0.0 0.0
0.0 0.1 0.1 0.3 0.3
(3) Take the square of each deviation.
0.09 cm2 0.04 0.04 0.00 0.00
0.00 0.01 0.01 0.09 0.09
(4) Average together all the squared deviations.
average = 0.04 cm2
(5) Take the square root. This is the standard devi-
ation.
standard deviation = 0.2 cm
If we’re using the symbol x for the length of the
couch, then the result for the length of the couch
would be stated as x = 203.1 ± 0.2 cm, or x = 203.1
cm and σx = 0.2 cm. Since the Greek letter sigma
(σ) is used as a symbol for the standard deviation, a
standard deviation is often referred to as “a sigma.”
Step (3) may seem somewhat mysterious. Why not
just skip it? Well, if you just went straight from
step (2) to step (4), taking a plain old average of
the deviations, you would find that the average is
zero! The positive and negative deviations always
cancel out exactly. Of course, you could just take
absolute values instead of squaring the deviations.
The main advantage of doing it the way I’ve outlined
above are that it is a standard method, so people will
know how you got the answer. (Another advantage
is that the standard deviation as I’ve described it
has certain nice mathematical properties.)
A common mistake when using the standard devi-
ation technique is to take too few measurements.
For instance, someone might take only two measure-
ments of the length of the sofa, and get 203.4 cm
and 203.4 cm. They would then infer a standard de-
viation of zero, which would be unrealistically small
because the two measurements happened to come
out the same.
In the following material, I’ll use the term “stan-
dard deviation” as a synonym for “error bar,” but
that does not imply that you must always use the
standard deviation method rather than the guessing
method or the 2/3 rule.
There is a utility on the class’s web page for calcu-
lating standard deviations.
Probability of deviations
You can see that although 0.2 cm is a good figure
for the typical size of the deviations of the mea-
surements of the length of the sofa from the aver-
age, some of the deviations are bigger and some are
smaller. Experience has shown that the following
probability estimates tend to hold true for how fre-
quently deviations of various sizes occur:
< 1 standard deviation about 2 times out of 3
1-2 standard deviations about 1 time out of 4
2-3 standard deviations about 1 time out of 20
3-4 standard deviations about 1 in 500
4-5 standard deviations about 1 in 16,000
> 5 standard deviations about 1 in 1,700,000
The probability of various sizes of deviations, shown
graphically. Areas under the bell curve correspond to
probabilities. For example, the probability that the mea-
surement will deviate from the truth by less than one stan-
dard deviation (±1σ) is about 34 × 2 = 68%, or about 2
out of 3. (J. Kemp, P. Strandmark, Wikipedia.)
Example: How significant?
In 1999, astronomers Webb et al. claimed to have
found evidence that the strength of electrical forces in
the ancient universe, soon after the big bang, was slightly
weaker than it is today. If correct, this would be the first
example ever discovered in which the laws of physics
changed over time. The difference was very small, 5.7±
1.0 parts per million, but still highly statistically signifi-
cant. Dividing, we get (5.7 − 0)/1.0 = 5.7 for the num-
ber of standard deviations by which their measurement
was different from the expected result of zero. Looking
at the table above, we see that if the true value really
was zero, the chances of this happening would be less
than one in a million. In general, five standard devia-
tions (“five sigma”) is considered the gold standard for
statistical significance.
However, there is a twist to this story that shows how
statistics always have to be taken with a grain of salt.
167
 In 2004, Chand et al. redid the measurement by a
more precise technique, and found that the change was
0.6 ± 0.6 parts per million. This is only one standard
deviation away from the expected value of 0, which
should be interpreted as being statistically consistent
with zero. If you measure something, and you think
you know what the result is supposed to be theoret-
ically, then one standard deviation is the amount you
typically expect to be off by — that’s why it’s called
the “standard” deviation. Moreover, the Chand result
is wildly statistically inconsistent with the Webb result
(see the example on page 171), which means that one
experiment or the other is a mistake. Most likely Webb
at al. underestimated their random errors, or perhaps
there were systematic errors in their experiment that
they didn’t realize were there.

Precision of an average
We decided that the standard deviation of our mea-
surements of the length of the couch was 0.2 cm,
i.e., the precision of each individual measurement
was about 0.2 cm. But I told you that the average,
203.1 cm, was more precise than any individual mea-
surement. How precise is the average? The answer
is that the standard deviation of the average equals
standard deviation of one measurement
√ .
number of measurements
(An example on page 170 gives the reasoning that
leads to the square root.) That means that you can
theoretically measure anything to any desired preci-
sion, simply by averaging together enough measure-
ments. In reality, no matter how small you make
your random error, you can’t get rid of systematic er-
rors by averaging, so after a while it becomes point-
less to take any more measurements.

168 Lab Appendix 2: Basic Error Analysis

169
 Appendix 3: Propagation of Errors
Propagation of the error from a
single variable
In the previous appendix we looked at techniques
for estimating the random errors of raw data, but
now we need to know how to evaluate the effects of
those random errors on a final result calculated from
the raw data. For instance, suppose you are given a
cube made of some unknown material, and you are
asked to determine its density. Density is defined
as ρ = m/v (ρ is the Greek letter “rho”), and the
volume of a cube with edges of length b is v = b3 , so
the formula
ρ = m/b3
will give you the density if you measure the cube’s
mass and the length of its sides. Suppose you mea-
sure the mass very accurately as m = 1.658±0.003 g,
but you know b = 0.85±0.06 cm with only two digits
of precision. Your best value for ρ is 1.658 g/(0.85 cm)3 =
2.7 g/cm3 .
How can you figure out how precise this value for ρ
is? We’ve already made sure not to keep more than
twosignificant figures for ρ, since the less accurate
piece of raw data had only two significant figures.
We expect the last significant figure to be somewhat
uncertain, but we don’t yet know how uncertain. A
simple method for this type of situation is simply to
change the raw data by one sigma, recalculate the
result, and see how much of a change occurred. In
this example, we add 0.06 cm to b for comparison.
b = 0.85 cm gave ρ = 2.7 g/cm3
b = 0.91 cm gives ρ = 2.0 g/cm3
The resulting change in the density was 0.7 g/cm3 ,
so that is our estimate for how much it could have
been off by:
ρ = 2.7 ± 0.7 g/cm3 . ence c of a guinea pig’s waist to be 10 cm, Using the
Propagation of the error from sev-
eral variables
What about the more general case in which no one
piece of raw data is clearly the main source of error?
For instance, suppose we get a more accurate mea-
surement of the edge of the cube, b = 0.851 ± 0.001
cm. In percentage terms, the accuracies of m and
b are roughly comparable, so both can cause sig-
170 Lab Appendix 3: Propagation of Errors
nificant errors in the density. The following more
general method can be applied in such cases:
(1) Change one of the raw measurements, say m, by
one standard deviation, and see by how much the
final result, ρ, changes. Use the symbol Qm for the
absolute value of that change.
m = 1.658 g gave ρ = 2.690 g/cm3
m = 1.661 g gives ρ = 2.695 g/cm3
Qm = change in ρ = 0.005 g/cm3
(2) Repeat step (1) for the other raw measurements.
b = 0.851 cm gave ρ = 2.690 g/cm3
b = 0.852 cm gives ρ = 2.681 g/cm3
Qb = change in ρ = 0.009 g/cm3
(3) The error bars on ρ are given by the formula
q
σ ρ = Q2m + Q2b ,
yielding σρ = 0.01 g/cm3 . Intuitively, the idea here
is that if our result could be off by an amount Qm
because of an error in m, and by Qb because of b,
then if the two errors were in the same direction, we
might by off by roughly |Qm | + |Qb |. However, it’s
equally likely that the two errors would be in oppo-
site directions,
p and at least partially cancel. The ex-
pression Q2m + Q2b gives an answer that’s smaller
than Qm + Qb , representing the fact that the cancel-
lation might happen.
The final result is ρ = 2.69 ± 0.01 g/cm3 .
Example: An average
On page 168 I claimed that averaging a bunch of mea-
surements reduces the error bars by the square root of
the number of measurements. We can now see that
this is a special case of propagation of errors.
For example, suppose Alice measures the circumfer-
guess method, she estimates that her error bars are
about ±1 cm (worse than the normal normal ∼ 1 mm
error bars for a tape measure, because the guinea pig
was squirming). Bob then measures the same thing,
and gets 12 cm. The average is computed as
A+B
c = ,
2
where A is Alice’s measurement, and B is Bob’s, giving
11 cm. If Alice had been off by one standard devia-
tion (1 cm), it would have changed the average by 0.5
cm, so we have QA = 0.5 cm, and likewisep QB = 0.5
cm. Combining these, we find σc = QA2 + QB2 = 0.7
√
cm, which is simply (1.0 cm)/ 2. The final result is
c = (11.0 ± 0.7) cm. (This violates the usual rule for
significant figures, which is that the final result should
have no more sig figs than the least precise piece of
data that went into the calculation. That’s okay, be-
cause the sig fig rules are just a quick and dirty way
of doing propagation of errors. We’ve done real propa-
gation of errors in this example, and it turns out that the
error is in the first decimal place, so the 0 in that place
is entitled to hold its head high as a real sig fig, albeit a
relatively imprecise one with an uncertainty of ±7.)
Example: The difference between two measurements
In the example on page 167, we saw that two groups
of scientists measured the same thing, and the results
were W = 5.7 ± 1.0 for Webb et al. and C = 0.6 ± 0.6
for Chand et al. It’s of interest to know whether the
difference between their two results is small enough to
be explained by random errors, or so big that it couldn’t
possibly have happened by chance, indicating that some-
one messed up. The figure shows each group’s results,
with error bars, on the number line. We see that the two
sets of error bars don’t overlap with one another, but er-
ror bars are not absolute limits, so it’s perfectly possible
to have non-overlapping error bars by chance, but the
gap between the error bars is very large compared to
the error bars themselves, so it looks implausible that
the results could be statistically consistent with one an-
other. I’ve tried to suggest this visually with the shading
underneath the data-points.

To get a sharper statistical test, we can calculate the

difference d between the two results,

d =W −C ,

which is 5.1. Since the operation is simply the subtrac-

tion of the two numbers, an error in either input simply
causes an error in the output that is of the same size.
Therefore pwe have QW = 1.0 and QC = 0.6, resulting
2
in σd = QW + QC2 = 1.2. We find that the difference
between the two results is d = 5.1 ± 1.2, which differs
from zero by 5.1/1.2 ≈ 4 standard deviations. Looking
at the table on page 167, we see that the chances that
d would be this big by chance are extremely small, less
than about one in ten thousand. We can conclude to a
high level of statistical confidence that the two groups’
measurements are inconsistent with one another, and
that one group is simply wrong.

171
 Appendix 4: Graphing
Review of Graphing
Many of your analyses will involve making graphs.
A graph can be an efficient way of presenting data
visually, assuming you include all the information
needed by the reader to interpret it. That means
labeling the axes and indicating the units in paren-
theses, as in the example. A title is also helpful.
Make sure that distances along the axes correctly
represent the differences in the quantity being plot-
ted. In the example, it would not have been correct
to space the points evenly in the horizontal direction,
because they were not actually measured at equally
spaced points in time.
Graphing on a Computer
Making graphs by hand in your lab notebook is fine,
but in some cases you may find it saves you time to
do graphs on a computer. For computer graphing,
I recommend OpenOffice, which is free, open-source
software. It’s installed on the computers in rooms
416 and 418. Because OpenOffice is free, you can
download it and put it on your own computer at
home without paying money. If you already know
Excel, it’s very similar — you almost can’t tell it’s
a different program.
Here’s a brief rundown on using OpenOffice:
On Windows, go to the Start menu and choose
Programs, OpenOffice.org, and Calc. On Linux,
do Applications, Office, OpenOffice.org, Spread-
sheet.
172 Lab Appendix 4: Graphing
Type in your x values in the first column, and
your y values in the second column. For sci-
entific notation, do, e.g., 5.2e-7 to represent
5.2 × 10−7 .
Select those two columns using the mouse.
From the Insert menu, do Chart.
When it offers you various styles of graphs to
choose from, choose the icon that shows a scat-
ter plot, with dots on it (XY Chart).
Adjust the scales so the actual data on the plot
is as big as possible, eliminating wasted space.
To do this, right-click anywhere on the axis,
choose the Scale tab, uncheck Automatic, and
put in the lower and upper limits you want.
Fitting a Straight Line to a Graph
by Hand
Often in this course you will end up graphing some
data points, fitting a straight line through them with
a ruler, and extracting the slope.
In this example, panel (a) shows the data, with error
bars on each data point. Panel (b) shows a best
fit, drawn by eye with a ruler. The slope of this
best fit line is 100 cm/s. Note that the slope should
be extracted from the line itself, not from two data
points. The line is more reliable than any pair of
individual data points.
In panel (c), a “worst believable fit” line has been
drawn, which is as different in slope as possible from
the best fit, while still pretty much staying consis-
tent the data (going through or close to most of the
error bars). Its slope is 60 cm/s. We can therefore
estimate that the precision of our slope is +40 cm/s.
There is a tendency when drawing a “worst believ-
able fit” line to draw instead an “unbelievably crazy
fit” line, as in panel (d). The line in panel (d), with
a very small slope, is just not believable compared
to the data — it is several standard deviations away
from most of the data points.
 data points. A dialog box will come up. Select the
Statistics tab, and under “Regression curves,” se-
lect the icon showing a line being fit to some data.
This will cause the line to be drawn on your graph.
To display the equation of the line, double-click on
the graph so that it’s surrounded by a gray border;
then right-click on the line, and do Insert Regres-
sion Curve Equation. By default your equation will
only have the slope and y-intercept shown with three
sig figs; if you need more precision, double-click on
the graph so it’s outlined in gray, right-click on the
equation, do Object Properties, Numbers, Scientific,
and add more zeroes after the decimal place under
Format code.
How accurate is your slope? A method for getting
error bars on the slope is to artificially change one
of your data points to reflect your estimate of how
much it could have been off, and then redo the fit
and find the new slope. The change in the slope
tells you the error in the slope that results from the
error in this data-point. You can then repeat this
for the other points and proceed as in appendix 3.
In some cases, such as the absolute zero lab and the
photoelectric effect lab, it’s very hard to tell how
accurate your raw data are a priori ; in these labs,
you can use the typical amount of deviation of the
points from the line as an estimate of their accuracy.

Fitting a Straight Line to a Graph

on a Computer
It’s also possible to fit a straight line to a graph using
computer software such as OpenOffice. E.g., lab 47
(the hydrogen atom) is a high-precision lab, and it’s
not possible to get a sufficiently accurate result by
hand.
To do this in OpenOffice, double-click on one of your

173
 Appendix 5: Finding Power Laws from Data
For many people, it is hard to imagine how scientists
originally came up with all the equations that can
now be found in textbooks. This appendix explains
one method for finding equations to describe data
from an experiment.
Linear and nonlinear relationships
When two variables x and y are related by an equa-
tion of the form
y = cx ,
where c is a constant (does not depend on x or y),
we say that a linear relationship exists between x
and y. As an example, a harp has many strings of
different lengths which are all of the same thickness
and made of the same material. If the mass of a
string is m and its length is L, then the equation
m = cL
will hold, where c is the mass per unit length, with
units of kg/m. Many quantities in the physical world
are instead related in a nonlinear fashion, i.e., the
relationship does not fit the above definition of lin-
earity. For instance, the mass of a steel ball bearing
is related to its diameter by an equation of the form
m = cd3 , The difference between each successive value of h
where c is the mass per unit volume, or density, of
steel. Doubling the diameter does not double the
mass, it increases it by a factor of eight.
It’s fairly easy to figure out what’s going on just
by staring at the numbers a little. Every time you
increase the height of the animal by a factor of 10, its
food consumption goes up by a factor of 100. This
implies that f must be proportional to the square of
h, or, displaying the proportionality constant k = 3
explicitly,
f = 3h2 .
Use of logarithms
Now we have found c = 3 and p = 2 by inspection,
but that would be much more difficult to do if these
weren’t all round numbers. A more generally appli-
cable method to use when you suspect a power-law
relationship is to take logarithms of both variables.
It doesn’t matter at all what base you use, as long as
you use the same base for both variables. Since the
data above were increasing by powers of 10, we’ll use
logarithms to the base 10, but personally I usually
just use natural logs for this kind of thing.
log10 h log10 f
shrew 0 0.48
rat 1 2.48
capybara 2 4.48
This is a big improvement, because differences are
so much simpler to work mentally with than ratios.
is 1, while f increases by 2 units each time. The
fact that the logs of the f 0 s increase twice as quickly
is the same as saying that f is proportional to the
square of h.

Power laws Log-log plots

Both examples above are of the general mathemati- Even better, the logarithms can be interpreted visu-
cal form ally using a graph, as shown on the next page. The
y = cxp , slope of this type of log-log graph gives the power
p. Although it is also possible to extract the pro-
which is known as a power law. In the case of a portionality constant, c, from such a graph, the pro-
linear relationship, p = 1. Consider the (made-up) portionality constant is usually much less interesting
experimental data shown in the table. than p. For instance, we would suspect that if p = 2
h=height of rodent f =food eaten per for rodents, then it might also equal 2 for frogs or
at the shoulder day (g) ants. Also, p would be the same regardless of what
(cm) units we used to measure the variables. The con-
shrew 1 3 stant c, however, would be different if we used dif-
rat 10 300 ferent units, and would also probably be different for
capybara 100 30,000 other types of animals.

174 Lab Appendix 5: Finding Power Laws from Data

175
 Appendix 6: Using the Photogate
The photogate
The photogate is a U-shaped thing about 10 cm
across, with an invisible infrared beam going across
the gap of the U, like the infrared beam of a TV re-
mote control. When something blocks the beam, an
electrical signal is sent through a wire to the com-
puter. We will use the photogate by sending moving
objects through it. The computer tells you for how
long the photogate was blocked, allowing you to cal-
culate the speed of the object as it passed through.
Plug the photogate into the DG1 plug on the inter-
face box.
Using the data
Often you may find that the software rounds off too
severely. For instance, when you’re in the mode for
measuring how long the photogate was blocked, you
want more than the three decimal places it offers by
default in the Delta-T column. To fix this, double-
click on the title of the Delta-T column, and select
a greater number of significant figures.

Using the software

If you’re using the ULI interface (beige box), use
Logger Pro 2, and make sure the interface box is
turned on before you boot up the computer.
If you’re using the USB interface, use Logger Pro 3.
From the Start menu at the lower left corner of the
screen, run Logger Pro (in Programs>Vernier Soft-
ware). It asks for permission to scan for the right
port — say OK. (If it complains that it can’t find the
port, you may be able to fix the problem if you quit
Logger Pro, power the interface off and on again,
and then get back in Logger Pro and try again.)
The next step depends on what mode you are using
the software in.

Using the software in different modes

For various labs, there will be three different modes
in which we’ll use the software. From the File menu,
do Open, and locate the file you need:

Mode for measuring how long the photogate

was blocked: Probes & Sensors > Photogate
> One Gate Timer
Mode for measuring the time between two in-
terruptions of the photogate: ...
Mode for measuring the period of a pendulum:
Probes & Sensors > Photogate > Pendulum

If there is no button for collecting data, it’s because

the interface box wasn’t turned on when you booted
up. Reboot.

176 Lab Appendix 6: Using the Photogate

177
 Appendix 7: Using a Multimeter
The most convenient instrument for measuring cur-
rents and voltage differences is called a digital mul-
timeter (DMM), or simply a multimeter. “Digital”
means that it shows the thing being measured on a
calculator-style LCD display. “Multimeter” means
that it can measure current, voltage, or resistance,
depending on how you have it set up. Since we have
many different types of multimeters, these instruc-
tions only cover the standard rules and methods that
apply to all such meters. You may need to check with
your instructor regarding a few of the particulars for
the meter you have available.
Blowing a fuse is not a big deal.
If you hook up your multimeter incorrectly, it is pos-
sible to blow a fuse inside. This is especially likely to
happen if you set up the meter to measure current
(meaning it has a small internal resistance) but hook
it up in parallel with a resistor, creating a large volt-
age difference across it. Blowing a fuse is not a big
problem, but it can be frustrating if you don’t real-
ize what’s happened. If your meter suddenly stops
working, you should check the fuse.

Measuring current
When using a meter to measure current, the meter
must be in series with the circuit, so that every elec-
tron going by is forced to go through the meter and
contribute to a current in the meter. Many multime-
ters have more than one scale for measuring a given
thing. For instance, a meter may have a milliamp
scale and an amp scale. One is used for measuring
small currents and the other for large currents. You
may not be sure in advance what scale is appropri-
ate, but that’s not big problem — once everything
is hooked up, you can try different scales and see
what’s appropriate. Use the switch or buttons on the
front to select one of the current scales. The connec-
tions to the meter should be made at the “common”
socket (“COM”) and at the socket labeled “A” for
Amperes.

Measuring voltage
For a voltage measurement, use the switch or but-
tons on the front to select one of the voltage scales.
(If you forget, and hook up the meter while the
switch is still on a current scale, you may blow a
fuse.) You always measure voltage differences with
a meter. One wire connects the meter to one point
in the circuit, and the other connects the meter to
another point in a circuit. The meter measures the
difference in voltage between those two points. For
example, to measure the voltage across a resistor,
you must put the meter in parallel with the resis-
tor. The connections to the meter should be made
at the “common” socket (“COM”) and at the socket
labeled “V” for Volts.

178 Lab Appendix 7: Using a Multimeter

179
 Appendix 8: High Voltage Safety Checklist
Name:
Never work with high voltages by yourself.
Do not leave HV wires exposed - make sure
there is insulation.
Turn the high-voltage supply off while working
on the circuit.
When the voltage is on, avoid using both hands
at once to touch the apparatus. Keep one hand in
your pocket while using the other to touch the ap-
paratus. That way, it is unlikely that you will get a
shock across your chest.
It is possible for an electric current to cause
your hand to clench involuntarily. If you observe this
happening to your partner, do not try to pry their
hand away, because you could become incapacitated
as well — simply turn off the switch or pull the plug
out of the wall.

180 Lab Appendix 8: High Voltage Safety Checklist

181
 Appendix 9: Laser Safety Checklist
Name:
Before beginning a lab using lasers, make sure you
understand these points, initial them, and show your
safety checklist to your instructor. If you don’t un-
derstand something, ask your instructor for clarifi-
cation.
The laser can damage your eyesight perma-
nently if the beam goes in your eye.
When you’re not using the laser, turn it off or
put something in front of it.
Keep it below eye level and keep the beam hor-
izontal. Don’t bend or squat so that your eye is near
the level of the beam.
Keep the beam confined to your own lab bench.
Whenever possible, orient your setup so that the
beam is going toward the wall. If the beam is going
to go off of your lab bench, use a backpack or a box
to block the beam.
Don’t let the beam hit shiny surfaces such as
faucets, because unpredictable reflections can result.

182 Lab Appendix 9: Laser Safety Checklist

183
 Appendix 10: The Open Publication License
Copyright (c) 1999-2001 B. Crowell and V. Roundy.
All rights reserved.
These materials are open-content licensed under the
OPL 1.0 license. A copy of the license is given below,
and the original is available at http://opencontent.org.
LICENSE
Terms and Conditions for Copying, Distributing, and
Modifying
Items other than copying, distributing, and modi-
fying the Content with which this license was dis-
tributed (such as using, etc.) are outside the scope
of this license.
1. You may copy and distribute exact replicas of the
OpenContent (OC) as you receive it, in any medium,
provided that you conspicuously and appropriately
publish on each copy an appropriate copyright no-
tice and disclaimer of warranty; keep intact all the
notices that refer to this License and to the absence
of any warranty; and give any other recipients of
the OC a copy of this License along with the OC.
You may at your option charge a fee for the media
and/or handling involved in creating a unique copy
of the OC for use offline, you may at your option
offer instructional support for the OC in exchange
for a fee, or you may at your option offer warranty
in exchange for a fee. You may not charge a fee for
the OC itself. You may not charge a fee for the sole
service of providing access to and/or use of the OC
via a network (e.g., the Internet), whether it be via
the world wide web, FTP, or any other method.
2. You may modify your copy or copies of the Open-
Content or any portion of it, thus forming works
based on the Content, and distribute such modifica-
tions or work under the terms of Section 1 above,
provided that you also meet all of these conditions:
a) You must cause the modified content to carry
prominent notices stating that you changed it, the
exact nature and content of the changes, and the
date of any change.
b) You must cause any work that you distribute or
publish, that in whole or in part contains or is de-
rived from the OC or any part thereof, to be licensed
as a whole at no charge to all third parties under the
terms of this License, unless otherwise permitted un-
der applicable Fair Use law. These requirements ap-
ply to the modified work as a whole. If identifiable
184 Lab Appendix 10: The Open Publication License
sections of that work are not derived from the OC,
and can be reasonably considered independent and
separate works in themselves, then this License, and
its terms, do not apply to those sections when you
distribute them as separate works. But when you
distribute the same sections as part of a whole which
is a work based on the OC, the distribution of the
whole must be on the terms of this License, whose
permissions for other licensees extend to the entire
whole, and thus to each and every part regardless
of who wrote it. Exceptions are made to this re-
quirement to release modified works free of charge
under this license only in compliance with Fair Use
law where applicable.
3. You are not required to accept this License, since
you have not signed it. However, nothing else grants
you permission to copy, distribute or modify the OC.
These actions are prohibited by law if you do not
accept this License. Therefore, by distributing or
translating the OC, or by deriving works herefrom,
you indicate your acceptance of this License to do
so, and all its terms and conditions for copying, dis-
tributing or translating the OC.
NO WARRANTY
4. BECAUSE THE OPENCONTENT (OC) IS LI-
CENSED FREE OF CHARGE, THERE IS NO WAR-
RANTY FOR THE OC, TO THE EXTENT PER-
MITTED BY APPLICABLE LAW. EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPY-
RIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE OC “AS IS” WITHOUT WAR-
RANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF MERCHANTABIL-
ITY AND FITNESS FOR A PARTICULAR PUR-
POSE. THE ENTIRE RISK OF USE OF THE OC
IS WITH YOU. SHOULD THE OC PROVE FAULTY,
INACCURATE, OR OTHERWISE UNACCEPTABLE
YOU ASSUME THE COST OF ALL NECESSARY
REPAIR OR CORRECTION.
5. IN NO EVENT UNLESS REQUIRED BY AP-
PLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER
PARTY WHO MAY MIRROR AND/OR REDIS-
TRIBUTE THE OC AS PERMITTED ABOVE, BE
LIABLE TO YOU FOR DAMAGES, INCLUDING
ANY GENERAL, SPECIAL, INCIDENTAL OR CON-
SEQUENTIAL DAMAGES ARISING OUT OF THE
USE OR INABILITY TO USE THE OC, EVEN IF
SUCH HOLDER OR OTHER PARTY HAS BEEN
ADVISED OF THE POSSIBILITY OF SUCH DAM-
AGES.

185