Physics Laboratory Handbook

Level 4 Handbook
Practical Physics Laboratory

2014/15
Sep
Oct
Oct
Oct
Oct
Nov
Nov
Nov
Nov
Dec
Dec
Dec
Jan
Feb
Feb
Feb
Feb
Mar
Mar
Mar
Mar
Apr
Apr
May
Week
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
23
30
7
29
5
12
19
26
5
12
19
26
Thu
2
9
16
23
30
6
13
20
27
4
11
18
C
M
M
K
K
O
O
M
K
K
O
O
Q1
C
Q2
C
C
M
I
G
F
F
R2
R1
Q2
46
I
G
F
F
R1
R2
Q1
13
C
M
M
K
Q2
C
R2
Q1
I
G
O
O
F
F
R1
79
C
M
M
K
Q1
C
R1
Q2
I
G
O
O
F
F
R2
1012
Pairs:
1618 1921 2224 2527 2830 3133
No labs - introductory lectures
Registration 11 a.m. 12 p.m., Data Analysis 2 p.m. 4 p.m.
I
I
I
I
G
G
G
G
G
G
G
I
I
I
K
K
M
M
C
C
Q1
K
K
M
M
C
C
Q2
O
O
K
K
M
M
C
O
O
K
K
M
M
C
F
F
O
O
K
K
M
Catch-up missed work & write formal reports
F
F
O
O
K
K
M
R1
R2
F
F
O
O
K
Christmas Holiday
R2
R1
F
F
O
O
K
Q1
Q2
R1
R2
F
F
O
Catch-up & write formal reports
Q2
Q1
R2
R1
F
F
O
C
C
Q1
Q2
R1
R2
F
C
C
Q2
Q1
R2
R1
F
M
M
C
C
Q1
Q2
R1
Catch-up & write formal reports
M
M
C
C
Q2
Q1
R2
Easter Holiday
Electronics Workshop
1315
Formal Reports: Formal reports are due in by 12.30 p.m. on the Monday of weeks 11, 16 and 21.
21
28
5
27
3
10
17
24
3
10
17
24
Tue
30
7
14
21
28
4
11
18
25
2
9
16
Catch-up Weeks: Lab attendance is compulsory. If you miss a session you will be required to explain your absence
and attend in the following catch-up week to complete the experiment.
20
27
4
26
2
9
16
23
2
9
16
23
Mon
29
6
13
20
27
3
10
17
24
1
8
15
Level 4 Practical Laboratory Timetable: 2014 2015
Laboratory classes are held on the days indicated, from 11 a.m. 1 p.m. and 2 4 p.m.
R1
O
F
F
R2
K
O
M
K
G
I
Q2
Q1
C
C
M
3436
K
O
O
F
M
K
C
M
G
I
R1
R2
Q1
Q2
C
3739
K
O
O
F
M
K
C
M
G
I
R2
R1
Q2
Q1
C
4042
Contents
I
A Guide to the Level 4 Teaching Laboratory
Code of Practice for the Teaching Laboratories
1 Introduction to the Level 4 Practical Physics Laboratory

1.1 Laboratory Hours . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Laboratory Staff . . . . . . . . . . . . . . . . . . . . . . . .
1.4 Aims of the Level 4 Lab . . . . . . . . . . . . . . . . . . . .
1.5 Laboratory Requirements . . . . . . . . . . . . . . . . . . .
1.6 Absences . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.7 Notes on Performing Experiments . . . . . . . . . . . . . .
1.8 Laboratory Notes and Written Reports . . . . . . . . . . . .
1.9 Errors and Data Analysis . . . . . . . . . . . . . . . . . . .
1.10 Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.11 Passing Requirements & Penalty System - Important! . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
11
11
11
12
13
15
16
16
17
17
18
2 A Guide to Good Laboratory Practice

2.1 Introduction . . . . . . . . . . . . . . .
2.2 Laboratory Notebooks . . . . . . . . .
2.3 Use of Graphs . . . . . . . . . . . . . .
2.4 Experimental Tips . . . . . . . . . . .
2.5 Risk Assessments . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
19
19
20
20
21
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
22
22
22
23
23
24
24
25
25
26
27
27
28
29
29
31
31
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3 Error Analysis: Important Concepts

3.1 Introduction . . . . . . . . . . . . . . . . . . .
3.2 Significance & Errors in Measurements . . . .
3.3 Random and Systematic Errors . . . . . . . .
3.4 Random errors in a series of readings . . . . .
3.5 Precision versus Accuracy . . . . . . . . . . .
3.6 Distribution of Measurements . . . . . . . . .
3.7 The Standard Error in a Single Observation .
3.8 The Standard Error in the Mean . . . . . . .
3.9 Obtaining and m from Experimental Data
3.10 Estimation of Errors . . . . . . . . . . . . . .
3.11 The Gaussian Error Distribution . . . . . . .
3.12 Combination of errors . . . . . . . . . . . . .
3.13 Errors in gradients of straight-line graphs . .
3.14 The method of least squares . . . . . . . . . .
Appendix A . . . . . . . . . . . . . . . . . . . . . .
Appendix B . . . . . . . . . . . . . . . . . . . . . .
3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
CONTENTS
4 How to Write Formal Reports

4.1 Introduction . . . . . . . . . .
4.2 Report Structure . . . . . . .
4.3 Marking . . . . . . . . . . . .
4.4 Plagiarism . . . . . . . . . . .
4.5 Formal Report Template . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
32
32
32
33
33
33
5 Mark Schemes
38
5.1 Practical Work Marking Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.2 Formal Report Marking Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.3 Explaining the 21-point marking scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
II
The Experiments
43
Measurement of the Speed of Light

45
Foucault method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Fibre-optic method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Optical Interference
52
Newtons Rings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Holography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
K The Black Body Radiation Law
60
M Measurement of the Charge of the Electron
68
74
Microwave Optics
Q1 Photoemission of Electrons from Metallic Surfaces
79
Q2 Charge:Mass Ratio of the Electron
85
R1 Thermionic Emission From Tungsten
92
R2 Heat Transfer
96
Part I
A Guide to the Level 4 Teaching

Laboratory
Safety Notice
One of the most important skills that you will learn in the Level 4 Physics laboratory is
how to do your experiments safely. Much of this skill is simply common sense. However, there are legal requirements regarding safety matters. The University is under an
obligation to provide adequate safety information and instructions. You, as students,
are under an obligation to follow safe procedures in the laboratory. Disregarding safety
can put yourself and others at risk of an accident or injury.
You will meet several potential hazards in the lab. These include chemicals, extremely
low and high temperatures, high voltages and laser light. However the risk of an
accident is very small if safe working procedures are followed. It is a very good
practice to take a few minutes during the planning of an experiment to assess possible
hazards and decide how to minimise the risk to yourself and others. This should, of
course, be done by consulting any safety information in the handbook and staff or
demonstrators as necessary.
It is important to remember that even if the experiment you are performing seems very
safe, you do not know the details of every other experiment that takes place in the
lab. Therefore everyone (including students, staff and visitors) must read and abide
by the procedures described in the Code of Practice in Section I.
If you have any further safety concerns, please raise them with laboratory staff or
Departmental Safety Officer. We hope you have an enjoyable and safe time in the
Level 4 Physics laboratory.
Code of Practice for the Teaching

Laboratories
Long hair must be tied back when instructed.
Good laboratory practice is essential if laboratories are to be safe places in which to work. The
safety measures described in this document should
be observed by all users of the teaching labs; students, demonstrators and staff. Users should be
aware that although their particular activity may
be deemed to involve no specific hazards there are
often other people performing experiments which
may be hazardous to all occupants of the lab.
Undergraduate students are not permitted entry to the teaching labs when a technician is not
present. Staff and postgraduate students should
inform a technician if they wish to use the teaching
labs outside of normal working hours.
Please note that the procedures detailed in this
document are general safety guidelines to be observed during all laboratory activities. Many experiments will have specific further hazards; you
will be required to read and sign a risk assessment
before performing those experiments.
Do not bring food or drinks into the lab. They

will be confiscated and may be thrown away.
Wash hands after leaving, especially before
eating or drinking.
Do not consume ice cubes or water from lab
taps.
Do not run in the lab or use MP3 players or
mobile phones.
Wall signs indicate regions where liquid nitrogen is frequently in use, and where bystanders
may be exposed to open laser beams. Avoid
these areas where possible.
Postgraduate demonstrators or technicians
should be consulted before unfamiliar equipment is turned on.
If a Darkroom In Use sign is activated,
knock on the door and wait for permission before entering.
Good Lab. Practice & Standard

Precautionary Measures
Report any breakages or spills to a technician

immediately to minimise the risk of injury or
slipping.
First, all undergraduate students must attend a

safety induction before they are permitted to attend a lab session. These inductions must be repeated annually. In addition to this, the following
rules should be observed:
Emergency Procedures
Contact details for first aiders are displayed near

Clothing must provide adequate protection most of the sinks in the teaching labs. In the case
from spillages and jewellery should be removed of accidents or other emergencies, the following prowhen instructed.
cedures should be followed.
Shoes must cover the whole foot to provide
some protection against crushing, chemical
burns, liquid nitrogen burns etc. Flip-flops,
sandals and ballet shoes are forbidden, and
shoes with significant heels are also not suitable due to the risk of slipping or tripping.
For electrical equipment, isolate the power at

the wall socket only if safe to do so. If you are
uncertain if it is safe, leave the equipment on.
In the case of an injury, call for a qualified first
aider.
If chemicals enter the eyes, eye wash stations
are available in rooms 2.14 and 2.18. Users
should locate the nearest eye wash station before using any chemicals. Inform a first aider
of the injury, and if possible provide details of
the chemicals involved.
Bags and coats should be stored in the white

cupboards provided to avoid obstructing the
aisles.
Wear lab coats, gloves, eye protection and any
other PPE when instructed.
8
9
Report all accidents or breakages to a lab technician immediately.
If the fire alarm sounds, evacuate immediately.
The fire doors will close but will remain unlocked. You will not need your U-card to leave
the lab.
where possible block the beams so that they

do not escape the immediate vicinity of the
experiment.
Areas where there may be open beams are indicated by wall signs and should be avoided by
non-essential persons.
Although the lasers are low power it is good
practice to remove all jewellery and tie long
hair back to avoid accidental reflections into
your (or someone elses) eyes.
Specific Hazards and Control

Measures identified within the
Teaching Laboratories
Liquid Nitrogen & Dry Ice - risk of burns and

asphyxiation in enclosed spaces.
The following items are all used frequently within

the teaching labs. If used incorrectly or carelessly
they can cause serious, potentially fatal, injuries.
You should familiarise yourself with the hazards
associated with each item and the procedures put in
place to minimise the risk to users and bystanders.
Wear PPE as instructed during use or transportation. Users will not be given access to
either liquid nitrogen or dry ice without receiving separate safety training. Bystanders
should avoid areas where liquid nitrogen or dry
ice is in use (indicated by wall signs).
Numerous extremely hot materials - sources

include boiling liquids & bright lamps. Risk
of burns, and lamps may release hazardous
vapours if damaged.
Label all containers clearly with the contents,
your name and a safety warning if required. Do
Check for damage before turning on lamps and
not move any unfamiliar or unlabelled chemireport any breakages to a technician immedicals or liquids - inform a technician. Wear PPE
ately. Do not use if damaged. Be aware that
as instructed and tie long hair back. Read risk
most lamps become very hot during use and
assessments thoroughly and ensure you know
the casing may also be hot enough to burn.
the emergency procedures for exposure before
Be very careful when transporting hot liquids,
using any chemicals.
and keep them in the plastic trays provided
whenever possible.
Compressed Gases - risk of asphyxiation, overRadioactive & X-ray sources - risk of burns
pressure explosion risk for apparatus.
and internal organ damage.
Cylinders should be operated by trained staff
Users will not be given access to radioactive
only; output pressure must be low ( 1 bar).
sources or keys to activate x-ray equipment
Electrical Appliances - risk of electrocution.
without first receiving separate safety training.
Do not turn on any electrical equipment, in- Working in Darkrooms - damage to persons or
cluding mains cables, if there is no in-date PAT
equipment due to trip hazards or unfamiliarity
sticker displayed. Request training before uswith the room.
ing any unfamiliar equipment. Many appliUsers should be familiar with the layout of
ances operate at high voltages so use shrouded
the room and the locations of light switches,
cables where indicated. Ask a trained user
and should minimise trip hazards by storing
to check the wiring before turning on any
all bags out of the pathways and all unused
high voltage equipment. Do not attempt to
stools under the benches.
take apart any electrical apparatus unless instructed to do so, and do not cover any elec- Working at height - not permitted without spetrical equipment.
cific safety training. Do not stand on benches
or stools to reach equipment - ask for help
Heavy movable equipment - risk of injury or
from someone who has undergone Working at
crushing if dropped or knocked.
Height training.
Chemicals - some toxic if ingested, danger of
chemical burns.
Lasers - risk of eye damage from various sources,

open beams across lab.
Waste Disposal
All teaching lab. lasers should be Class II so

the blink response should protect against eye Sharps should be disposed of in the appropriate
damage. Do not stare into the beams, and
(metal or glass) sharps bin by a technician.
10
Students should avoid touching sharp items
and should inform a demonstrator or technician of any breakages.
Waste chemicals will be disposed of by the
Chemical Technician, via the teaching lab
technicians.
Contaminated paper & gloves etc. should
be disposed of in the yellow chemicals bin in
darkroom 2.13a, or via the technician if the
Darkroom in Use sign is lit.
When leaving the labs

At the end of each day you should:
Tidy your bench and return all equipment
to the condition it was in when you arrived,
putting any rubbish into the bins, or chemical
waste bin if necessary.
Turn off all electrical equipment and switch off
sockets at the mains (including computers!).
Wash your hands before leaving, or immediately after leaving, the lab.
Version 2, created 30/07/13.
Chapter 1
Introduction to the Level 4 Practical

Physics Laboratory
This guide describes the roles of the staff you will
meet in the teaching laboratory, and gives tips on
good experimental techniques, methods for calculating errors and writing reports. It details the laboratory requirements, submission procedures and
mark schemes, and describes the steps to be taken
if you are unable to attend a laboratory session.
The information contained in this guide was correct at the time of printing, however procedures
may occasionally be altered throughout the year. If
any changes are made, you will be notified via email
and Blackboard. Experiment scripts and any updated versions of this handbook will also be available on Blackboard.
features within the experimental apparatus. To do

so constitutes a serious University disciplinary offence, which may lead to dismissal from the course.
If you find any cause for concern with the safety of
the apparatus, you must report it immediately to
the laboratory technicians.
1.1
1.3.1
Laboratory Hours
The laboratory timetable is shown at the front of

this handbook. For laboratory work, the first year
is divided into three groups, each attending on either Monday, Tuesday or Thursday. The laboratory sessions run between 11 a.m. - 1 p.m. and 2
- 4 p.m. on each of these days. You should arrive
on time and complete all experimental work within
these hours. The laboratory is closed between 1
and 2 p.m. every day.
1.2
1.3
Laboratory Staff
The Laboratory Coordinator is Dr. Jaap Velthuis

([email protected], room 4.48). He is responsible for the overall supervision and development of the first year physics laboratory.
Laboratory Technicians
Dr. Gemma Winter and the assistant lab technicians (Nic and Sanghamitra) are responsible for
the day-to-day running of the laboratory, including
safety supervision, the supply and repair of apparatus and the issue of instruments (e.g. calipers,
torches and stopwatches).
Any accidents or breakages of apparatus should
be reported to the technicians immediately.
They also keep a stock of consumables and record
your attendance, marks and submission of work.
Safety
The Physics Teaching Laboratories are governed by 1.3.2 Staff Demonstrators

the same safety rules as any other working laboratory. You must observe the rules in the Code of The teaching and marking of work in the laboratory is actively overseen by a member of the acaPractice at all times.
Students wearing unsuitable clothing or shoes demic staff on each of the laboratory days. They
(e.g. flip-flops, shoes with significant heels) will be are present in the laboratory to advise on the teachrequired to wear a lab coat or asked to leave the lab. ing of experimental techniques and skills, to moderTo decide if your clothes and shoes are suitable, ask ate the marking of demonstrators and to provide a
yourself if they offer protection from spillages and second opinion when experimental difficulties arise.
This year the staff demonstrators are:
falling objects.
The experiments in the laboratory have been
Mondays
Dr. Monica Berry
carefully assessed for safety. Under no circumTuesdays
Dr. Sven Friedemann
stances should you attempt to interfere with safety
Thursdays Prof. Rob Richardson
11
12
CHAPTER 1. INTRODUCTION TO THE LEVEL 4 PRACTICAL PHYSICS LABORATORY
The staff demonstrators are happy to help with

The strategy behind practical work is fundamenexperimental issues, questions about marking or tally different from the approach in lectures. In the
more general queries about the lab so you should lab you dont just learn a set of abstract theories
not feel uncomfortable approaching them.
and facts; you must also contend with the vagaries
and imperfections of the real world. You must learn
to recognize the resulting problems and deal with
1.3.3 Postgraduate Demonstrators
them without compromising your aims. You are
These are research students who, under the super- expected to learn to think for yourself and then act
vision of the staff demonstrator, are responsible logically and practically on this basis.
In addition to these general remarks, it is worthfor the teaching of physics and experimental techniques in the lab. They will be present in the lab while to identify specific skills that you should acquire or improve as a result of your years work
when you have to attend.
You will have the same demonstrator until the in the Level 4 Lab. Some of these are listed below.
end of week 6, then after that each demonstrator All of the experiments involve several of these skills,
specialises in supervising one or two of the main lab although the balance differs between experiments.
experiments and remains in charge of these. You
will be supervised and marked by several different
demonstrators throughout the year.
1.4.1 Organizational Skills
We encourage you to discuss challenges with the
To adopt a systematic approach to testing hycurrent experiment you are performing with the
potheses and trouble-shooting.
specialist demonstrator in charge of that experiment, and general challenges and issues with your
To be able to plan and execute systematic meapersonal demonstrator.
surements to test a theoretical relationship and
Your personal demonstrator can also advise you
to determine the relevant constants in that reif any other problems are affecting your ability to
lationship.
work, but you do not have to speak to them about
To be able to estimate and minimize the erthis - for any serious or personal issues you may
rors in quantities determined and to establish
speak to the Lab Coordinator or the First Year Cotheir combined effect on the final result. To
ordinator, or you can ask for a confidential meeting
be aware of systematic errors that bias your
with the Senior Tutor.
result and to test for and then correct them.
Specialist Demonstrators: These supervise one
To plan, test and then carry out an effecor two main experiments throughout the year,
tive programme of experiments to investigate
and no more than 6 pairs of students at any
a physical system where the theoretical betime. Do not be afraid to approach the demonhaviour is not initially known.
strator for advice about your experiment. One
of the most effective ways of learning ex To be able to extract as much information as
perimental physics is through discussion with
possible from an experiment and produce relisomeone already actively involved in research.
able quantitative results that you can defend.
Personal Demonstrators: On your first lab day
you will be assigned to a particular demonstra- 1.4.2 Communication Skills
tor. They will supervise your introductory ex To keep an accurate and clear written record of
periments and your first full experiment, and
your experiments, from which you can recover
they will act as a mentor for the rest of the
appropriate information, even months later.
year, keeping track of your marks and speaking to you to make sure there are no prob To interact constructively and fruitfully with
lems. If you are unhappy with the work, or do
others: partners, demonstrators, staff etc.
not understand any comments made by your
other demonstrators then do not be afraid to
To write clear, well-argued and concise formal
ask your personal demonstrator for help.
reports.
1.4
Aims of the Level 4 Lab
A primary aim of your work in the Level 4 Lab is to

begin to acquire the qualities needed by a professional physicist: critical sense, logical analysis and
the ability to control a practical situation.
1.4.3
Experimental Skills
To be able to set up and use basic measurement kit (e.g. oscilloscope, microscope);
To be able to set up and use simple optical
systems with lenses and mirrors;
1.5. LABORATORY REQUIREMENTS
13
To be able to wire up a given circuit so that it Pre-lab Assignments: Short tests (accessed
functions properly;
through Blackboard) that must be passed
before you attend the lab to begin an ex To develop a knowledge of basic electronic
periment. They are intended to help you
components, their functions and their appliprepare for experiments and consolidate your
cation in experimental physics;
knowledge of the theory. Read the experiment
script thoroughly before you attempt them.
To be able to use computers to process and
Some tests are relatively simple and may be
display data, as well as using word-processing
completed quickly, others may take a couple
packages;
of hours.
To be able to analyse data, perform correct erThe whole quiz must be completed before you
ror analysis, present results in an appropriate
attend the lab to begin an experiment or late
fashion and to make quantitative comparisons
penalties will be applied and you may not be
with other measurements.
permitted access to the lab.
Finally, the physics lab is where you should begin to develop your own creative powers whilst remaining in control of a physical situation. In the
first year you are presented with fairly simple experiments. These will give you results but you will
need to think carefully and critically to achieve excellent results. The challenges you need to overcome will mean you are well prepared for the more
open-ended experiments next year and for project
work in your final year.
1.5
Laboratory Requirements
Each quiz may be attempted twice - a second

attempt is compulsory if you fail the first time.
If you take the quiz twice your highest mark
will be recorded. The pass mark is 60%.
Some pre-lab tests contribute to your end-ofyear grade and some do not, but in all cases
you are expected to make a satisfactory attempt at completing them. Students who have
not achieved 60% in the quiz before attending
the lab will be given an additional assignment
by the technicians. This assignment must be
completed to a satisfactory standard before the
student is permitted to start the experiment.
The level 4 lab consists of 19 experimental sessions Experiments: There are 7 main experiments,
each of which takes two weeks to complete,
and 6 lectures covering error analysis and formal
arranged in a cycle through Weeks 5-21. (Exreport writing. You are required to attend ALL
periments Q and R are each made up of 2
sessions and must sign the attendance register each
1-week experiments.) To complete the experweek. During the year you will be required to comiments you will need to undertake careful explete and submit various assignments as follows:
perimental measurements in the lab followed
by a thorough analysis and interpretation of
Data Analysis & Errors: You will be given a
your results.
data analysis assignment early in the Autumn
term, based on the material in the lab lectures Formal Reports: You will be required to write 3
and the data analysis session in week 2. You
formal reports during the year: they are due
must submit this assignment to pass the lab
in weeks 11, 16 and 21.
- students who receive marks below 70% are
You must achieve a pass mark in the first readvised to ask for guidance and then repeat it.
port to pass the laboratory. If you fail to
These skills are fundamental for your lab work
achieve a pass mark on this report you will
throughout the year.
be required to rewrite it.
Experiments I & G: Introductory workshops
The marks for the second and third reports
which must be completed to a satisfactory
contribute to your final mark for the lab.
standard to pass the first year lab.
Electronics Workshop: The application of elecFor experiment I you should make notes in
tronics is an important tool in a modern lab.
your lab book which your demonstrator must
In the final few weeks of the course you will
sign at the end of the session to indicate that
undertake the electronics workshop where you
your work was satisfactory, but the work does
will learn about and build simple circuits. This
not need to be handed in.
will also provide you with an opportunity to
Experiment G will be marked; if you fail to
apply the mathematics of complex numbers to
obtain a pass mark you will have to repeat the
a.c. circuit theory. The electronics workshop is
experiment. The mark for the assignment will
not assessed but you are required to attend unnot however be included as part of the end of
til the staff are satisfied by your understanding
year assessment of the lab.
and progress.
14
Introductory Assignments:
Errors Exercise
Experiment I
Pre-lab G
Experiment G
Pre-lab Assignments:
C
F
K
M
O
Q1
Q2
R1
R2
Experiments:
C
F
K
M
O
Q1
Q2
R1
R2
Formal Reports:
1 - week 11
2 - week 16
3 - week 21
% of Grade
Notes
Advised to repeat until 70% attained

Must pass - extra sessions until completed
Advised to repeat until 70% attained
6/20 needed (before penalties) - repeat if required.
2
2
2
2
2
1
1
1
1
60%
60%
60%
60%
60%
60%
60%
60%
60%
needed
needed
needed
needed
needed
needed
needed
needed
needed
2
2
2
2
2
2
2
2
2
attempts
attempts
attempts
attempts
attempts
attempts
attempts
attempts
attempts
permitted
permitted
permitted
permitted
permitted
permitted
permitted
permitted
permitted
8
8
8
8
8
4
4
4
4
15
15
Must pass - Edit until 6/20 attained
Must pass - extra sessions until completed
Table 1.1: The assessed components of the first year Practical Laboratory, their contribution to the final
mark, and the passing requirements.
The final mark recorded for your Laboratory unit

will be determined as shown in Table 1.1, and the
minimum requirements for passing the course are
described in detail in section 1.11.
All summative work is assessed according to the
university guidelines. A mark greater than or equal
to 70% will be regarded as a first class mark. Between 60% and 70% is an upper-second class mark;
between 50% and 60% is a lower-second class mark;
between 40% and 50% will be regarded as a third
class mark, and less than 40% is a fail. The mark
schemes are described in chapter 5.
A passing mark is not required for each individual contributing experiment but if you regularly receive failing grades you will be in danger of failing
the Practical Laboratory course and should seek
some advice from a demonstrator or staff member.
Students are not normally permitted to repeat
experiments that contribute to the final mark.
We understand that some students may experience difficulties throughout the year which make
attending labs or achieving deadlines difficult. In

these situations it is crucial that you read and follow the advice given in section 1.6.
1.5.1
Submitting Laboratory Work
Pre-lab work
For safety reasons, the pre-lab. assignment must be
completed before you attend the laboratory. You
will not be permitted to start the experiment until
you have completed it and penalties will be subtracted from your final mark for late submission.
The deadline for completing the Pre-lab tests is
10.30a.m. on your lab day.
Experimental lab books

At the end of each day you should make sure you
dont leave the lab until your demonstrator has
1.6. ABSENCES
signed each page of your lab book. Work that is
not signed will not be marked.
The formal deadline is one day after you finish
the experiment, at 4.00 p.m., (i.e., if you finish
an experiment at 4 p.m. on Tuesday in Week 5,
your book is due in at 4 p.m. on the Wednesday of
Week 5).
In this time you should complete the data analysis and then hand your notebook in for marking.
Any work completed outside of the lab (i.e., after
the final demonstrators signature) must be completed independently, not with your lab partner.
Loose pieces of paper will be thrown away
and will not be marked.
Lab books must be placed in the red lockers outside the south doors of the main lab
before the deadline, or strict penalties for late
submission will be applied (see section 1.11). Books
must not be handed directly to the technicians or
the demonstrators.
Note: Lab books for experiments C, F, K, M
and O should be submitted after you have completed both lab sessions. Experiments R1, R2, Q1
and Q2 should be submitted after each lab day
since the two parts of the experiment are completely independent.
Your lab books will be returned to you at the
start of your lab session the following week. During
this lab session you will be allocated a time for your
assessment. You should take your book to your
marker at this time - it is your responsibility to be
on time.
If you are confused about these deadlines please
ask a staff member to explain the late penalties
can make a huge difference to your final mark!
1.5.2
Submitting Formal Reports
All formal reports must be submitted electronically

via Blackboard in Word or Adobe pdf format. If
you are not sure how to do this, ask your demonstrator for advice. The deadlines for electronic submission of the three formal reports are 12:30 p.m.
on the Mondays of Weeks 11, 16 and 21.
You must also print a hard copy and hand it in
to the technicians before 4 p.m. on the same day.
The demonstrators always mark the printed copy
so you must proof-read!
The electronically recorded submission
time will be used as the definitive submission time. As with lab books, penalties will be
applied for late submission of formal reports.
When these reports are returned to you a few
weeks later you should read through the comments
and the feedback sheet and ask the marker, staff
demonstrator, or your personal demonstrator if you
15
do not understand any points. We strongly encourage you to proactively talk to them, as any feedback you get is intended to aid you in improving
your scientific report writing capabilities.
1.5.3
Plagiarism - Important!
All written work will be checked electronically and

manually for evidence of plagiarism. To plagiarise anothers work (including your lab partners)
is considered to be a very serious offence and may
lead to expulsion from the University. For further
information see the level 4 handbook.
To ensure students are performing the experiments correctly and obtaining their own data, your
lab books must be signed by the demonstrator of
the experiment before you leave the lab at the end
of each session. Both partners in a pair must write
all data in their own lab books in pen, or the
demonstrator will not sign your books. Any data
that is missing a signature will not be marked.
If you are entering your data directly into a computer spreadsheet, you must print out your data
and glue it into your lab book, and the demonstrator will sign the printout.
1.6
Absences
Attendance at all laboratory sessions is mandatory,

however you may be unable to attend on a particular day. The actions you need to take in these
circumstances are given below (see also the level 4
handbook).
Catch-up sessions are scheduled throughout the
year - if you miss an experiment you will be expected to complete it during the catch-up week (if
not before). There will not normally be demonstrators to help you in these weeks.
1.6.1
Illness
If you are absent from the Department due to illness for a period of 5 days or less you must complete
the University self-certification form which can be
downloaded from:
http://www.bristol.ac.uk/science/
undergraduates/
and hand it directly to the Undergraduate Administration team in room 3.20 within 2 working days
of the end of the period of absence. If possible, you
should also inform the lab technicians before your
lab session. If you record multiple absences in this
way you will be asked for an explanation from the
Senior Tutor, Dr. Barnes, who will decide on the
appropriate action to be taken.
16
If you are absent for a period of more than 5

days, or if you miss a formal report deadline due
to illness, you must bring or send the extenuating
circumstances form (also available from the above
web page) with a medical certificate from your doctor to the Senior Tutor or the Student Administration Manager, Ms. Barb Perks. They can keep the
contents of these forms confidential, and after assessing the effect of this illness on your work they
will advise the lab staff of any action (e.g. extensions) to be taken.
1.6.2
Absence due to other serious

mitigating circumstances
In the case of absence for other serious reasons

(e.g. a close family bereavement) you should see
your year co-ordinator or the Senior Tutor - see the
Physics Student Handbook for more details. Absences that are deemed reasonable may be treated
in a similar manner to self-certified illnesses, or you
may be asked to complete an Extenuating Circumstances form. If necessary, alternative laboratory
work will be arranged.
1.6.3
Pre-arranged absence
In all other cases you are expected to attend all

of your lab sessions. In exceptional circumstances
you may request, by prior arrangement, an agreed
absence from the laboratory. If agreed, you will
be required to complete the lab session you missed.
Note: taking holidays during term time is
unacceptable and will not be sanctioned.
1.6.4
Unauthorised absences
Before you arrive in the lab to start an experiment, you must spend some time planning
it and thinking about safety. The mandatory
pre-lab tests are aimed to help you develop this
process.
Decide what you will have to measure, which
readings you will need to take and in what order they are best taken.
If possible, decide over what ranges your measurements should be taken and under what
conditions, so that the final results are as complete and accurate as possible. You may not
know enough about the apparatus to do this
before you actually begin. You should be able
to learn enough from your first readings to proceed immediately in the right direction, or you
may have to make some preliminary measurements before starting the main experiment.
Plot readings as they are taken on a rough
graph in your lab book and use this to decide
what to do next. Examination of the results
while the experiment is in progress will almost
always save you time and often allow you to
spot errors in the experimental setup.
Avoid the uncritical accumulation of pages of
data. Think about what you are trying to
achieve you could get more useful information from fewer well-chosen measurements.
1.8
Laboratory Notes
Written Reports
and
Keeping a good record in your lab notebook is as

important as doing the experiment itself. You will
depend on these notes for the detailed analysis of
your data and the writing of formal reports.
The degree of formality to be adopted depends
for whom the account is intended, for example,
whether it is solely for your own information or is to
be circulated within a research team or submitted
for publication in a learned scientific journal.
Absences that are not excused through the completion of either self-certification or extenuating
circumstances forms will be treated as unauthorised. Unauthorised absence from a laboratory
session will mean a mark of zero being recorded
for that experiment and may lead to failure of the
laboratory and hence your first year. You will still
be expected to complete the experiment to a satis- Experimental notes
factory standard.
Your lab notebook is not expected to be particularly formal, although it is expected to be a madescription of your lab work. It is important
1.7 Notes on Performing Ex- ture
to develop the skill of good note taking during experiments
periments notes should be complete yet concise.
This is essential for writing subsequent formal reStudents are expected to work in pairs for experi- ports. Your demonstrators will give you guidance
ments. The principal aim of the experiments is to on keeping a good notebook. Chapter 2 gives furintroduce you to experimental methods. Here are ther guidance.
some suggestions that will help you to get the most
The data analysis should be started during the
out of each experiment:
lab session but you may have to complete it at
1.10. ELECTRONICS
home. The data analysis, final graphs and error
analysis are assessed and a brief discussion of the
results and summary of your conclusions are also
expected. You are expected to perform the data
analysis and discuss the results independently.
Formal Reports
Throughout the year you are expected to write
three formal reports of your experiments, as though
for publication in a scientific journal e.g. Physical
Review B. This is an important skill to acquire;
journals insist that articles are short and concise
yet convey all information necessary for the reader
to appraise the results and conclusions. The reports will be assessed by a demonstrator and moderated by a staff demonstrator. You may choose
which experiment to write up, subject to some restrictions, so you are advised to discuss your choice
with your personal demonstrator. Guidance in formal report writing will be given in the lab lectures
in the Autumn term (also Chapter 4).
1.9
Errors and Data Analysis
The results of all measurements are to some extent

uncertain, even if actual mistakes are absent. This
unavoidable uncertainty is termed the error of observation. Unless it is known how large the error is
likely to be, a measurement of a physical quantity
is of doubtful significance. Therefore the numerical
results of an experiment should always be accompanied by an estimate of their uncertainty.
The subject of errors will be discussed in the
lab lectures during the Autumn term. Chapter 3
contains further information. Also, the introductory experiment G and the associated pre-lab assignment are designed to teach the estimation and
treatment of errors. In all subsequent experiments,
you will be expected to perform an error analysis as
part of the routine treatment of your experimental
results. Several experiments include detailed instructions of how to perform a rigorous error analysis and careful assessment of your results.
1.10
Electronics
In addition to the Physics experiments in the lab

there is a short electronics workshop. The reason
for focusing on this particular branch of applied
physics is its immense practical importance. Almost all modern research experiments in physics
make some use of electronics and many would be
impossible without it. Unfamiliarity with basic
electronics is therefore a serious handicap for an
experimental physicist.
17
18
1.11
Passing Requirements & Penalty System - Important!
Passing the Level 4 Practical Laboratory Course

The normal minimum requirements for passing the Physics Laboratory are:
satisfactory attendance,
satisfactory completion of the Electronics Workshop and Experiment I,
completion and submission of experiments C, F, G, K, M, O, Q1, Q2, R1 and R2,
submission of 3 formal reports, and
an overall mark of 40 % or more.
All required submissions must be received by 11th May 2015 at the latest. Any required submissions
not received by this time will receive a mark of zero and will result in a fail mark for the lab, regardless
of the average mark.
Extreme circumstances may be discussed in confidence with the Senior Tutor or Ms. Barb Perks see the student handbook for further details.
Late submission of Laboratory Notebooks or Reports
It is essential that laboratory work be handed in on time. Demonstrators and staff set aside fixed
periods of time to carry out the assessment within their overall workload, and late submission
causes significant delay and disruption. For this reason penalties will be applied unless an extension
is agreed beforehand by the Senior Tutor, or he agrees to waive the penalties due to strong mitigating
circumstances.
Penalties will be applied according to the University regulations (See the Universitys Regulations
& Code of Practice for Taught Programmes).
Formal Reports and Experimental Lab Books: The mark given for the work will be reduced
by 3 marks (of the 20 available) for each day it is late, including weekends. Once seven days have
elapsed an assessment mark of zero will recorded. Note, however, submission of the work and a
satisfactory mark for the assessment will still be required for the credit points to be awarded. This
means that even if you know the work will have such severe penalties that you will receive a mark
of zero, you must still submit a satisfactory attempt in order to pass the lab.
Pre-lab Assignments: The mark given for the work will be reduced by 10 marks (of the 100
available) for each day it is late, including weekends. Once seven days have passed an assessment
mark of zero will be recorded.
Extenuating Circumstances
If there are extenuating circumstances, including illness, then these must be brought to the attention
of the Senior Tutor before the due date apart from situations when it would be impossible. All
illness must be certificated. If you are having difficulties coping with your studies it is vital that you
inform the department (see the level 4 handbook for who to contact) so that we can make suitable
allowances where necessary.
Unsatisfactory performance
It is your responsibility to keep track of your marks and progress. Your marks can be seen on the
Marks, Absences and Feedback (MAF) system. If you are concerned about your marks you should
speak to a staff member immediately.
Chapter 2
A Guide to Good Laboratory

Practice
2.1
Introduction
Chapter 1 mentions that you are expected to give

estimates of the accuracy of your experimental
results. The importance of doing so is obvious
enough: unless there is some idea of how large the
error in a measured quantity is likely to be, the
significance of the result is in doubt. Without an
estimate of error it is not possible to decide, for
example, whether one experimental result differs
significantly from another, or from the prediction
of some theory.
The next two chapters are concerned with all aspects of the treatment of data, and the reduction
of errors therein. In the first, you will be given an
outline of good practice in taking notes and recording experimental data. The ideas outlined here will
assist you in recording your results in a clear and
understandable manner. More will be said about
this in the first Lab Lecture. The following chapter, on the other hand, will take you through some
of the more important concepts in the estimation
and calculation of experimental errors.
2.2
Laboratory Notebooks
An essential part of any experiment is keeping a

record of everything that is done. This should be
a clear, concise description of the equipment used,
and the measurements taken. Your notes should
be sufficient that you can readily interpret them
after, say, one year has passed. The following are
considered good practice in maintaining a clear laboratory notebook:
1. Each entry in your lab book should begin with
the date and a title for the piece of work.
2. Record each measurement directly into the
book do not perform any mental arithmetic
on it.
19
3. You are not permitted to make rough records

on pieces of paper and copy them into your lab.
book at a later time. This is time consuming,
and prone to error. You should record everything in your lab. book, but note that the very
nature of experimental work means it may not
end up the most tidy of documents.
4. Make a careful note of the location and contents of any computer filenames containing
data that you need to analyse later. Make sure
these files are backed up (CD, DVD, USB stick
etc...) and note the location of the backups.
5. Use simple, annotated diagrams to indicate
which measurements are being taken.
6. Do not copy detailed instructions directly from
the lab script - it is unnecessary. If you wish to
record complicated instructions you may print
a copy of the lab script and glue the relevant
section into your lab book (stating the source).
7. When measurements are recorded, write down
clearly and unambiguously:
(a)
(b)
(c)
(d)
what the quantity is;

its measured value;
any available estimate of its error;
the units used.
8. Stick securely into your notebook any computer generated graphs along with a note of
how they were produced and the location of
the data on the computer.
9. Use tables to record data whenever possible.
Put units in the table headings and always try
to select a natural power of ten to work with,
avoiding too many zeros. For example, in the
following table, the Youngs modulus of iron is
2.11 1011 Nm2 .
Substance
Youngs Modulus (1011 Nm2 )
Iron
2.11
20
CHAPTER 2. A GUIDE TO GOOD LABORATORY PRACTICE
2.3
Use of Graphs
3. Lines or curves should be consistent with your

model do not simply join up the points:
Graphs are valuable for illustrating the relationship

between two variables, and should be used whenever necessary both in your rough notes in lab.
books and in your formal reports.
Young's Modulus (1011 Nm-2)
2.7
In your lab. book, you should give graphs a title

to explain the data that is presented, however when
writing a formal report you should put this information underneath the graph in a figure caption
instead.
There are a number of points to bear in mind
when drawing graphs:
2.4
2.3
2.2
2.1
2
2.7
GOOD
2.6
2.5
2.4
2.3
2.6
300
310
320
T (K)
330
340
350
320
T (K)
330
340
350
BAD
2.5
2.4
2.3
2.2
2.1
2
1.9
290
2.2
GOOD
2.5
1.9
290
1. Choose sensible axes, so that the data points

are well spread over the graph paper. i.e. like
this:
2.7
2.6
300
310
2.1
2
1.9
290
300
310
320
330
340
4. Show the error bars wherever possible - if they

are too small to see then say so. This is an
example of a good graph:
350
and not like this:
2.8
BAD
2.5
2
1.5
1
0.5
50
100
150
200
250
300
350
2. Label the axes clearly with the variables and

their units:
2.7
2.4
2.2
2
1.8
290
0
0
2.6
2.6
300
310
320
T (K)
330
340
350
It is good practise to draw rough graphs as you

collect data. This way, you quickly get an idea of
whether your measurements span the correct range
of values.
2.5
2.4
2.4
2.3
2.2
2.4.1
2.1
2
1.9
290
300
310
320
T (K)
330
340
350
Experimental Tips
Common Sense
There are a number of techniques you can apply

when performing experiments that will make the
operation run more smoothly and, hopefully, enable you to obtain more reliable results. Some of
2.5. RISK ASSESSMENTS
21
them are listed here. A more complete discussion for a 10 C temperature change, or 0.1 mm in 1 m.
can be found in [1].
Clearly, for the most accurate work, thermal expansion will make a significant contribution to the
1. After reading the script and completing the errors.
prelab, plan which measurements you intend
Be aware that the very ends of the wooden rules
to take. Estimate how long each part of the quickly become worn out you should not use the
experiment will take and plan your time ac- final mm for measurements of any kind.
cordingly.
For more accurate measurements, vernier callipers
and micrometers are available. Your demon2. It is always worth performing a preliminary
strator
will explain the operation of these to you.
experiment before you commence serious data
collection. This has the advantages that:
(a) You learn how to use the apparatus;
3.
4.
5.
6.
2.4.3
Reading Meters
(b) You establish that each piece of equip- Beware of parallax errors when reading analogue
ment is functioning correctly;
(needle) voltmeters and ammeters. Always read
(c) You find a suitable range for each variable the meter with your eyes directly above the scale.
Also, beware of the apparent accuracy of digital
in the experiment;
meters.
These may have many digits of precision,
(d) You can estimate the errors in the difbut
unless
they have been properly calibrated (a job
ferent quantities in the experiment, and
normally
performed
by a specialist instrumentation
decide which will have the most influence
workshop)
you
may
not be justified in quoting a
on your results.
reading to more than 3 significant figures.
Review your results as you go, to ensure you
collect data over the correct range of values.
Risk Assessments
Think about the numbers you are recording. 2.5
Are they roughly what you were expecting?
Most experiment scripts contain warnings of any
Plot simple graphs where appropriate.
hazards associated with the experiment. This is
Assess the main sources of error at the start of one reason that you must read the scripts thorthe experiment, then concentrate on reducing oughly before you arrive in the lab. You should
the largest errors (e.g. by repeating measure- think about these hazards and try to plan your exments, or taking extra care with readings).
periment so that it minimises any risks.
When you arrive in the lab you may be given a
Repeat measurements to reduce mistakes and
copy of the School of Physics risk assessment for
provide a means of estimating errors.
your experiment. You must read it thoroughly and
Be aware of experimental drift and system- sign it to indicate your understanding before you
atic variations. These may occur as temper- are permitted to use any equipment. In future years
atures change during the day, or as electronic you will be asked to write these yourself before you
equipment warms up. In particular, the val- can start an experiment.
ues of resistors may change when they pass
large currents. Plan your experiments to try to
minimise these effects, e.g. by randomising the Bibliography
sequence of measurements, or repeating measurements at widely spaced time intervals.
[1] Squires, G.L., Practical Physics, 4th. Ed., Cambridge University Press (2001).
2.4.2
Measuring lengths
The Level 4 Physics laboratory has metre rules

made of both wood and steel. The wooden rules
should be used for most measurements. However,
they could be in error by as much as 0.5 mm over
their lengths, and so should be calibrated against
the steel rules. It should be possible to make measurements accurate to about 0.2 mm using this
method. The steel rules have been calibrated at
25 C. The thermal expansion of steel results in a
change in length of approximately 1 part in 104
Chapter 3
Error Analysis: Important Concepts

3.1
Introduction
Learning how to analyse experimental errors is

probably the most important skill that you will derive from the Level 4 Physics laboratory. Consequently, we place a great deal of emphasis on its
teaching, and will be testing your understanding
in a variety of ways (e.g. problem sheets, Maple
TA assignments, rigorous data analyses for some
experiments, and exams) throughout the year.
You should discuss the ideas and procedures outlined here with your demonstrator, and apply the
methods to all experimental write-ups and formal
reports from now on. You will be expected to be
able to perform this level of data analysis routinely
in the Level 5 Physics Laboratory and beyond. You
will learn more advanced data analysis techniques
as you progress through the Physics course.
For further information on data treatment and
basic techniques in experimental physics in general,
the following book is recommended:
A Practical Guide to Data Analysis for Physical Science Students by Louis Lyons (Cambridge University Press).
You may also wish to look at:
An Introduction to Error Analysis by J.R.
Taylor (University Science Books);
Experimental Measurements Precision and
Truth by N.C. Barford (John Wiley & Sons);
Practical Physics, 4th. Ed. by G.L. Squires
(Cambridge University Press, 2001).
do not expect the value to be exactly equal to the

true value. We need to give some estimate of how
close the value we measured is likely to be to the
true value. Suppose we measure the focal length of
a lens as 128 1 mm. This is a statement of the
probability or likelihood that the focal length lies
between 127 and 129 mm. It does not mean that
we are certain it lies between these limits, but just
that there is a probability that it does.
Experimental results are usually quoted in the
form F F where F is the best value obtainable
from the measurement and F is the standard error in it. Note that F is not the maximum possible error; the limits F are only intended to
indicate how large the error is likely to be. Even
when such a maximum possible error is known
(and usually it is not) it does not provide a very
satisfactory measure of the accuracy. For example,
it is almost always more useful to know that the
true value of a quantity has a 68 % chance of being
between 9.9 and 10.1, or perhaps a 99.7 % chance
of being between 9.7 and 10.3, than to know that it
cannot possibly be less than 5 or greater than 15.
If we measure some quantity under different conditions and the values we measure are different, we
often want to know if this difference is significant or
not. For example, we want to know whether temperature has an effect on the electrical resistance of
copper wire. So we measure its resistance at two
temperatures:
200.025
200.034
at 10 C
at 20 C
Much of the material in this chapter was derived Is the difference between these two values significant? We cannot say without knowing the errors
from the book by Squires.
in each case. If the error in each value is 0.001 ,
the difference is significant, whereas if the error is
3.2 Significance & Errors in 0.01 , then it is not significant.
In designing an experiment, it is important to
Measurements
consider all sources of error that could contribute
Estimating errors is important. If we do not do so, to the error in the final result. We must know how
we cannot draw significant conclusions from exper- large a final error we can tolerate for the result to be
imental results. When we measure a quantity, we significant in terms of the aims of the experiment.
22
3.4. RANDOM ERRORS IN A SERIES OF READINGS
23
For example, two rival theories for the resistivity of as a result his value for e was:
copper might predict different temperature dependence for the resistivity. To test the two theories
e = (1.591 0.002) 1019 C
experimentally, the error in the measured change of
resistivity with temperature must be smaller than The currently recommended value2 is:
the difference between the predictions of the two
theories.
e = (1.602176487 0.000000040) 1019 C
The final value in an experiment is usually arrived at by combining several primary measured
quantities. Errors in the primary quantities contribute to the error in the final result. In general,
each primary error contributes to a different extent,
and so to reduce the final error in an experiment
it is important to know which primary errors contribute most to the final error and concentrate on
reducing these first. Later in this chapter, we will
describe how to combine the estimated errors in
primary measurements to give the probable error
in a result calculated from them.
3.3
Random and Systematic

Errors
A random error is defined as one that does not affect the mean of an infinite series of repeated measurements, whereas a systematic error displaces it
from its true value. The classification of a particular error may depend on what we include in
our series of determinations. For example, let us
suppose that the resistances of a group of supposedly 100 resistors actually range from 99 to 101 ,
with mean value 100 . If one of them is accepted
at its face value when used as a standard resistance in a Wheatstone bridge1 , an error possibly
as great as 1 % will be introduced into any measurements taken. This error will enter systematically
into a series of readings taken with the same bridge;
but if a series is taken using a different standard resistor each time, the same error may be regarded
as random.
The significance of systematic errors cannot be
over-emphasised. They are inherent in any experimental design, and much of the art of experimentation concerns devising methods of eliminating them. Yet, because systematic errors are specific to each experimental arrangement, no general
guidance can be given for their treatment. Even
the most famous of experimenters have been badly
caught out by systematic errors in the past. A classic example is Millikans oil-drop experiment for
measuring the charge on an electron (experiment
M). The method depends on knowing the viscosity
of air. Millikan took a value that was too low, and
1 A Wheatstone bridge is used to compare two voltages
or resistances (see Experiment E1 ).
which is well outside Millikans estimate for random errors. Up until 1930, several other physical
constants, such as Plancks constant h and Avogadros number NA were derived from Millikans
value for e, and were consequently in error by more
than 0.5 %.
The reduction both of systematic and of random
errors, by improvement of the apparatus and perhaps of the measuring technique, is an important
matter of experimental design. Our concern here,
however, is not with the reduction of errors, but
with their assessment. We consider how random
(but not, of course, systematic) errors can be estimated from the results of repeated measurements.
3.4
Random errors in a series

of readings
Suppose that x1 , x2 . . . xn are n readings of a quantity whose measurement is subject to random error.
We shall assume, as is usually (but not invariably)
the case, that an error of a given size is as likely
to be positive as negative. The mean of these n
measurements is:
n
x=
1X
xi
n i=1
(3.1)
For example, supposing the measurement of resistance of the copper wire was made 8 times (n = 8)
and the values were:
1)
2)
3)
4)
4.615
4.638
4.597
4.634
5)
6)
7)
8)
4.613
4.623
4.659
4.623
then, using equation (3.1), x = 4.625 .

We need some quantity that gives a measure of
the spread of the 8 values about the mean. The
narrower the spread, the more likely the mean is
expected to be to the true value (ignoring systematic errors):
2 CODATA 2006, Taylor and Newell, Rev. Mod. Phys.
80, 633
24
CHAPTER 3. ERROR ANALYSIS: IMPORTANT CONCEPTS
smaller
spread:
x
x
larger
spread:
Number of
readings/interval
4.6
4.7
R/
x
To define this spread, we need to introduce the Figure 3.1: Histogram of 8 measurements of resisidea of distribution. However, first we will digress tance.
slightly to discuss some important terminology.
f(x )
3.5
Precision versus Accuracy
There is a tendency to use precision and accuracy

interchangeably. In the context of errors, however,
there is an important distinction between the two.
We use the term precise to indicate a result that
is relatively free from random error, and accurate
to indicate one that is free from systematic error.
Thus, a measurement might be accurate, but not
very precise, which means that it is essentially correct, but we do not have many digits of precision
in the value. Conversely, a value might be precise,
but not very accurate. A classic example would
be a digital volt meter which had 7 digits of precision, but which had an unknown zero offset error,
rendering most of the digits of no value. It is a
common mistake to believe the readings on digital
instruments, just because they are digital. It is always wise to check the accuracy of any instrument
claiming such precision.
3.6
Distribution of Measurements
<x>
=X
Figure 3.2: The resultant histogram when the number of measurements tends to infinity. The mean of
the distribution, hxi, tends to the true value, X.
we can say that the probability summed over the
whole distribution is:
Z
f (x)dx = 1
i.e. the distribution is normalised, and the mean of

distribution is:
Z
hxi =
xf (x)dx
We may represent any set of measurements by a There are two extremely important concepts to
histogram. For example, the 8 measurements of re- grasp at this point.
sistance above can be plotted on a histogram, here
1. The distribution shown in Fig. 3.2 would be
with intervals of 0.02 (see Fig. 3.6):
obtained if an infinite number of independent
With increasing number of measurements, the inmeasurements were performed. Therefore, we
tervals can be made smaller. As the number of meacan never hope to measure this distribution
surements tends to infinity, the distribution curve
exactly. However, the mean value of this disbecomes smooth (see Fig. 3.2).
tribution, hxi would be regarded as the true
In the distribution shown, the fraction of the tovalue, X, of the quantity being measured.
tal measurements in each interval is plotted. f (x)
is the distribution function, such that f (x)dx is the
2. Our actual set of measurements (for example,
fraction of the total readings that lie in the interour 8 measurements of resistance) is a random
val x to x + dx. This is the same as saying that
sample taken from this distribution. It will
f (x)dx is the probability that a single measurement
have a mean value, x , that differs from X by
taken at random will lie in this interval. From this
a random and unknown amount. It is the aim
3.8. THE STANDARD ERROR IN THE MEAN
25
of error analysis, to try to estimate the error

in x i.e. to estimate by how much x deviates
from the true value, X.
mean of 6
measurements
0.8
0.6
f(x)
In what follows, we will make frequent use of

two types of error: the standard error in a single
observation, and the standard error in the mean.
Definitions of these two quantities follow.
single
measurement
0.4
0.2
3.7
The Standard Error in a

Single Observation
The standard error in a single observation, , is

defined as the width, or spread, of the distribution
shown in Fig. 3.2. The experimental distribution is
inherent to the particular experiment and quantity
being measured. When we make a single measurement, it will lie somewhere within this distribution.
gives us a measure of the accuracy of that single
measurement.
The error, e, in a measured value, x, where X is
the true value is given by:
0
-3
-2
-1
0
x
Figure 3.3: A comparison of the distribution of single measurements (solid) with the distribution of
the means of groups of 6 measurements (dashed).
The distribution of means is narrower, indicating
that the mean is known more accurately as the
number of measurements increases.
it is this quantity that decreases as we increase the

number of independent measurements. We will denote it by m . In most cases we want to measure
e=xX
the mean and therefore the error on the mean is
the interesting number.
The root-mean-square (rms) value of e for all meaA set of n measurements results in a mean. If
surements is denoted by which is the standard
we measure q such sets each containing n measure2
deviation of the distribution. is known as the
ments, we will have q values of the mean which will
variance, and is defined by:
form
another distribution in themselves. This is
Z
best illustrated with a diagram.
2
2
2
= he i =
(x X) f (x)dx
In Fig. 3.3, we show two distributions: the true
distribution, obtained from an infinite number of

The standard deviation is a measure of the spread
single measurements, and a distribution formed
of the distribution:
from the means of sets of 6 readings. The distribution of the means is narrower than the initial distribution. We can obtain the standard deviation of
this distribution of means, which we will call m .
m can be regarded as a measure of the error in

the mean value of a single set of n measurements.
It is therefore known as the standard error in the
mean.
X
The relation between and m is given by:
small - precise data set
m =
n
This result is derived in Appendix A. The important point to note, is that m decreases as more
X
measurements are taken, i.e. as n increases.
large - imprecise data set
The difference between the standard deviation of
the distribution () and the standard error in the
(m ) is important, and so it is worth restat3.8 The Standard Error in the mean
ing it in a few words. The standard deviation, is
a measure of the distribution of the measured valMean
ues. This distribution does not become narrower
The standard error in the mean may be regarded as if more measurements are taken, it becomes betthe most important error to know about, because ter defined, i.e. the value of the standard deviation
26
does not decrease with more measurements, but

the distribution curve becomes smoother. On the
other hand, the value of the mean does become
known more accurately (i.e. m decreases) as more
measurements are taken.
3.9
Obtaining and m from

Experimental Data
have standard errors of / 2n and m / 2n, respectively. An estimate of the error obtained from
10 readings is therefore likely to be uncertain by
about 20 %, and one from 100 readings by 7 %. It
follows that it is seldom meaningful to state an error to more than one, or at most two, significant
figures. Throughout the Level 4 and 5 labs, you are
expected to quote your final error to one significant
figure, and hence your main result to the same level
of precision.
As seen above, the standard

error in the mean of n
observations, m , is 1/ n times the standard error

in a single observation, . However, can only
be obtained if we know the complete shape of the
single measurement distribution, which implies an
infinite number of measurements.
3.9.2
The best estimate of is:
v
u n
u1 X
=t
e2
n i=1 i
where ei is the error in the ith reading i.e.
Useful formulae
For a set of n measurements, xi :

the mean is:
e i = xi X
However, the errors come from knowing the true
value X, which we dont have. The way around
this difficulty is to use residuals, di , which are like
errors, but they are defined in terms of the mean of
the sample of n measurements, rather than X, i.e.:
x=
the variance is:

n
s2 =
d i = xi x
where we know both xi and x. The rms value of all
the residuals from a set of measurements is known
as the standard deviation of the sample, s, where:
n
s2 =
1X 2
1X
2
di =
(xi x)
n i=1
n i=1
Be sure to distinguish between the standard deviation of the sample, s, and the standard deviation
of the infinite distribution, they are not quite
the same. It is shown in Appendix B that we can
derive the standard error in the mean, m , and the
standard error in a single observation, , from our
experimental value of s:
r
n
s
and
=s
m =
n1
n1
3.9.1
1X
xi
n i=1
1X
2
(xi x)
n i=1
the standard error in a single measurement is:

v
u
r
n
u 1 X
n
2
=s
=t
(xi x)
n1
n 1 i=1
the standard error on the mean is:
The accuracy of an error esti- 3.9.3

mate
s
m = =
n
n1
Example calculation of the

standard error in the mean
The accuracy with which m can be estimated is

not great unless very large numbers of readings are
taken. In fact, it can be shown that the values The following table illustrates the calculation of
of and m obtained from n readings themselves and m from a set of 8 measurements of a quantity:
3.11. THE GAUSSIAN ERROR DISTRIBUTION

Observations: Deviations from mean:
xi
di = xi x
d2i
565
2.25
5.1
572
+4.75
22.6
563
4.25
18.1
567
0.25
0.1
571
+3.75
14.1
576
+8.75
76.6
558
9.25
85.8
566
1.25
1.6
4538
34.5
224.0
x = 567.25
4.3
s2 = 28.0
i
1
2
3
4
5
6
7
8
Sums:
Means:
ignoring
signs
and:
s = 28.0 = 5.3
p
= 8/7 5.3 = 6.1
m = 6.1/ 8 = 2.1
Therefore:
x = 567 2
We find:
and so:
3.10
Estimation of Errors
The calculation of standard errors is based on making many measurements of a quantity and assumes
that random events will cause the measured value
to be distributed about the true value. However, it
is not always possible or appropriate to take multiple measurements of a quantity, but we still require
some estimate of the error involved in the measurement. For example, you might measure a length
with a metre rule. You would not expect to repeat
such a measurement. In this case, careful consideration and estimation of the factors likely to influence the accuracy of the value must be made in
order to arrive at the best estimate of the likely
error.
Those errors that are important should be assessed, whenever possible, by repeating the affected
readings perhaps 5 or 10 times. It is only useful
to do this, however, if repetition reveals a genuine
spread in the results. In many simple cases the
readings may all be the same. This often happens
when the accuracy of a reading is limited by the
coarseness of the scale from which it is taken. Thus
if a steady indication of 44.7 scale divisions is read
to the nearest division, the reading is always 45.
An error of 0.3 divisions enters systematically into
every reading, and cannot be assessed or reduced
by repetition. The error introduced by reading to
the nearest division is random in the sense that it
is as likely to be positive as negative. Since the
error ranges from zero to half a division, the average error in such a measurement is a quarter, and
the standard error about a third, of a scale division.
These errors will apply no matter how many identical readings make up each measurement. Generally
27
in this case the error in the reading will be quoted
as 0.5of thesmallestscaledivision.
3.11
The Gaussian Error Distribution
So far we have assumed only that our random errors

are as likely to be positive as negative, and have
discussed how to estimate their r.m.s. values (standard errors). Some further assumption is needed
if we are to find the probability of an error of a
given size. The one usually made is that a random
error is itself the result of a large number of contributions, each as likely to be positive as negative.
It can then be shown that the probability P (y)dy
that a particular reading will have an error between
y and y + dy is:

y2
1
P (y)dy = exp 2 dy
2
2
where is the standard error ofone reading, as
discussed above. The factor 1/ 2 ensures that
the probability of the reading lying between
and + is 1, i.e., it is a normalising factor. The
function P (y) is sketched below, without the normalising factor (it is also tabulated in Table 3.1).
1
exp(-y /2 )
2
0.8
0.6
0.4
0.2
-3
-2
-1
0
y/
Such a distribution of errors is termed a normal

or Gaussian distribution. We might expect it to
apply best to errors containing a large number of independent and approximately equal contributions.
It should not be supposed that all random errors
necessarily distribute themselves in this way. There
is no obvious reason why, for example, errors arising from the use of a galvanometer of inadequate
sensitivity in a Wheatstone bridge should follow the
Gaussian law. Fortunately the errors in this and in
many other ordinary physical measurements do appear to obey it sufficiently closely for it to be useful
to apply to them certain relations that are strictly
correct only for a true Gaussian.
For a Gaussian distribution, the probability that
a particular reading lies within y0 of the true
28
A and Bi in B. If we were to make a large numTable 3.1: The Gaussian function, and the area
ber n of separate determinations of F , the standard
under the Gaussian curve, for various values of y/.
error F of a single determination would be given
by:
2
2
R +y0
y
y
1
exp 22 dy
y/ exp 22
1X
2 y0
Fi2
F 2 =
n
0
1
0
i
1X
0.2
0.980
0.159
2
(Ai + Bi )
=
n
0.4
0.923
0.311
i
0.6
0.835
0.452
1X
1X
=
A2i +
Bi2
0.8
0.726
0.576
n i
n i
1.0
0.606
0.683
2X
+
Ai Bi
1.2
0.487
0.770
n i
1.4
0.375
0.838
1.6
0.278
0.890
Since Ai Bi is as likely to be positive as negative,
1.8
0.198
0.928
the last term may be neglected and we obtain:
2.0
0.135
0.955
2.2
0.089
0.972
F 2 = A2 + B 2
2.4
0.056
0.984
2.6
0.034
0.991
3.12.1 Combining errors in real ex2.8
0.020
0.995
periments
3.0
0.0110
0.997
Several methods exist to combine errors. We insist
3.5
0.0022
0.9995
that you learn the partial differentiation method.
4.0
0.0003
0.99997
Partial differentiation is relatively simple. You differentiate with respect to one variable while treating the other variables as constants. A major advalue is:
vantage is that the method also works when the

Z y0
2
variables are not independent.
1
y
exp 2 dy
Here follows a brief explanation of the partial
2
2 y0
differentiation method. Suppose that you measure
This function also is tabulated in Table 3.1. It can the power radiated by the surface of a black body.
be seen that the odds are about 2 to 1 in favour This is given by the formula
of a reading lying within of the true value (the
P = AT 4
(3.2)
greyed region of the figure), and 20 to 1 on its being
within 2. There is only one chance in about
where P is the power radiated, A the area,
30,000 of a reading being wrong by more than 4,
the Stefan-Boltzmann constant and T the absolute
if the distribution is Gaussian.
temperature. As you can write equation 3.2 as
It must be remembered, however, that pracP = AT T T T
(3.3)
tical error distributions commonly depart most
markedly from the Gaussian law in the region when
y0 is large; i.e. large errors are often much more it is obvious that an uncertainty in T counts much
more than the same relative uncertainty in A. You
abundant than Table 3.1 suggests.
can understand this because
3.12
Combination of errors
It often happens that the result, F , of an experiment has to be calculated from the measured values of several independent quantities
A, B, C . . .. Suppose that we know the standard errors A, B, C . . . in the measurements of these,
and wish to find the standard error F in our final
result F .
As an example, suppose that F = A + B. The
actual error Fi in one determination of F is obviously equal to the sum of the actual errors Ai in
P
A
P
P
T
=
=
T 4
1
=
AT 4
A
4AT 3
4
=
AT 4
T
(3.4)
(3.5)
Hence a mistake in T counts four times as heavy as

a mistake in A.
In general we calculate the error in a value F
where F = f (pi ), where pi are the parameters in
the theory, using the formula
X f (pi ) 2
2
2

yi =
(3.6)
xi xi
i
3.14. THE METHOD OF LEAST SQUARES
29
As an example, suppose that F = X/Y then

By eye. Draw the lines with the maximum and
the minimum slope that still go through most
2

1
X
2
2
2
of the points. Then you can find the slope as
Y
(3.7)
F = 2 X +
Y
Y2
mmax mmin
mmax + mmin
or
(3.14)
m=
2
2

2
2 2
2
F
X Y
Y
X 2 + 2 4 Y 2 (3.8)
=
Taking points in pairs. This works best if the x
F
Y 2X 2
X Y

2
2
values are equally spaced. Assume there are 8
X
Y
=
+
(3.9)
points. Take the points in pairs, 1 with 5, 2
X
Y
with 6 and so on. For each pair of points, determine the gradient of the straight line joining
Another nice example is the lens equation. If you
them, m. Finally, calculate the mean gradient,
were to calculate the error in the focal length f
and its standard error, in the usual way. The
after measuring u and v you get
best line given by this method has a slope m,
2

uv
v
2
2
and passes through the point (x, y).
u (3.10)
f
=
u + v (u + v)2
2

Using the method of least squares. This is
u
uv
2
the best method to use. Good scientific data
v
+
u + v (u + v)2
analysis programs will have the option to
v4
u4
carry out such fits with the appropriate error
2
2
=
u +
v (3.11)
(u + v)4
(u + v)4
calculations (as well as more sophisticated
techniques). You will also encounter the 2
or to get the relative error
method. This is the same as the method of

2
least squares but every point is weighted by
f
(u + v)2
v4
=
u2 (3.12)
2
2
4
its error. At this stage a detailed discussion
f
u v
(u + v)
goes too far, but you should remember that a
2
4
(u + v)
u
2
+ 2 2
v
good fit has a 2 /n.d.f of approximately 1.
u v
(u + v)4
v2
u2 +
(3.13)
=
u2 (u + v)2
3.14 The method of least
u2
squares
v 2
v 2 (u + v)2
The method of least squares is a general technique used for fitting all manner of curves or analytical models to experimental data. It is very
widely used. We will outline the method here; for
further details, see the book by Squires.
The term least squares refers to a minimisation
of the squares of the deviations of the experimental data points from their ideal values. Let yT (pi )
be the expected value of your measurements. This
value will depend on a set of parameters pi . If we
assume that the x values are correct, then the deIf
B = 0.1, it is clear that F = viation in yi for a given xi will be:
A = 2 and
4 + 0.01 4. Thus, the smaller error, B,
contributes very little to the final error, and could
di = yi yT (i)
justifiably be ignored in this case. A similar analysis can be applied to each experimental situation. i.e. the experimental value minus the ideal value
Remember that your error is generally only stated predicted by eq. 3.16. If we sum the squares of the
to one significant figure.
deviations of all n of the datapoints, we obtain a
figure of merit for the fit, S:
From the examples shown, you can see that it
is not necessary to find all of the errors very accurately. There is no need to waste time finding the
individual error in some measured quantity if that
error obviously contributes very little to the error
in the final result.
For example, a variable, F , is given by F = A +
B. The standard error in F is given by:
q
2
2
F = (A) + (B)
3.13
Errors in gradients of
straight-line graphs
S=
n
X
d2i
i=1
There are three common methods of estimating the We take the squares of the deviations to avoid canerror in the gradient of a graph:
cellation of points above and below the line. The
30
aim of the exercise is now to minimise S, by vary- and:

ing the parameters pi the smaller the value of S,
c = y mx
(3.20)
the better our fit will be.
where x and y are the means of the x and y values
Minimisation proceeds by differentiating S with
of the datapoints, respectively.
respect to all parameters pi , and looking for staThe standard errors in gradient and intercept are
tionary values in the usual way. Thus:
given by:
Pn
n
X
1 i=1 d2i
2
2
(m) '
S =
(yi yT (pi ))
D (n 2)
i=1
S
pi
(3.15)
Strictly speaking these are partial derivatives, but

the mechanism of differentiating is no different from
what you have already met. (Partial differentiation will be treated in your Maths course later in
the year, and will be required by both the Oscillations and Waves course and the Fields course).
The expressions appear quite daunting because of
the summations. However, all we are doing is differentiating each term in the sum in turn.
3.14.1
(c)2 '
where:
D=
x2
1
+
n
D
i=1
3.14.2
Straight line through origin
In the special case that the line must pass through

the origin (c = 0), equation 3.19 becomes:
General straight line
where m and c are the gradient and intercept respectively and are to be determined. Any particular datapoint (xi , yi ) will not lie on this line but
will deviate from it. If we assume that the x values
are correct, then the deviation in yi for a given xi
will be:
di = yi (mxi + c)
Minimisation proceeds by differentiating S with
respect to both m and c,
S
n
X
(yi mxi c)2
i=1
S
m
= 2
S
c
= 2
n
X
xi (yi mxi c) = 0 (3.17)
i=1
n
X
(yi mxi c) = 0
(3.18)
i=1
Equations 3.17 and 3.18 may be solved simultaneously for m and c. Omitting a few lines of
algebra, we obtain:
n
X
(xi x)yi
m=
i=1
n
X
(xi x)
i=1
(3.19)
2
d2i
(n 2)
i=1
n
X
(xi x)2
n
X
Here is a worked example for a straight line. In the

case of a straight line we will write the straight line
as:
y = mx + c
(3.16)
Pn
m=
xi yi
i=1
n
X
x2i
i=1
and the standard error is:

2
Pn
d2i
2
i=1 xi (n 1)
1
(m) ' Pn
i=1
3.14. THE METHOD OF LEAST SQUARES
Appendix A
31
Appendix B
The error in the mean, E, is given by E = x X, Given:

where X is the true value. Therefore:
E =xX
1X
xi X
n i=1
1X
(xi X)
n
X
1
ei
n
=
=
=
xi x = e i E
Therefore:
s2
=
=
(3.21)
j6=i
This comes from:

X 2
ei
=
e i = xi X
We obtain:
and so:
1 X 2
1 XX
ei + 2
ei ej
E2 = 2
n i
n i
and
=
(e1 + e2 + e3 + . . . + en )
(e1 + e2 + e3 + . . . + en )
= e21 + e22 + e23 + . . .
+e1 e2 + e1 e3 + e2 e3 + . . .
1X
2
(xi x)
n
1X
2
(ei E)
n
1 X 2 2E X
ei
ei + E 2
n
n
1X 2
ei E 2
n
Taking the average over many sets of measurements, as we did before, we obtain the result:
2
2
s = 2 m
Using m = / n from Appendix A, we obtain:
1
2
n
2
This E is for one set of n measurements. If we
2
2 =
s
and
m
=
s
imagine that we take many such sets of n mean1
n1
surements, equation (3.21) will hold for each. By
2
adding all the equations and dividing by the num- The quantity s is not known, since we only obber P
of sets, we average over all the sets. The average tain one set of measurements. Therefore, the best
2
2
of ( ei ) is nhe2 i. The average of each term in the available estimate of it is s . Thus:
r
double sum is zero, because ei and ej are indepens
n
m '
and
's
dent and the average of each is zero. Therefore:
n1
n1
2
1 2
E = he i
We can therefore calculate both the standard ern
ror in the mean and the standard error in a single
But:
observation in terms of quantities that are known,

2
m
= E2
and 2 = he2 i
i.e. the standard deviation of the sample, s, and the
number of points in the sample, n.
and so:
m =
n
as previously stated.
Chapter 4
How to Write Formal Reports

4.1
Introduction
4.2
Report Structure
A full discussion of formal report writing will be Formal reports should be as long as necessary to
given in one of the laboratory lectures during the convey the information and no longer. In the Level
Autumn term. This chapter contains a brief in- 4 teaching labs a limit of 8 pages with a font size
troduction, followed by a Formal Report Template of 11 must be observed. These 8 pages should conwhich contains fuller explanations of the require- tain all of the sections listed below, including any
graphs or diagrams. This means that you must sements of each part of the report.
Your reports should be based on one of the lect your figures carefully to ensure all necessary
eight lab. experiments in this book (not G or (but no excess) information is included.
Appendices may be included in addition to the 8
Holography). Your lab. partner and you should,
where possible, choose different experiments. Be- page limit, however they should be no longer than 2
fore starting, discuss your choice of experiment pages in total. Appendices should be used for clarification of detailed analysis steps, complex derivawith your personal demonstrator.
The report should be aimed at a reader with tions, or occasionally descriptions of equipment pesimilar background knowledge to you but no prior culiarities, but all information necessary for underknowledge of the particular experiment. Good re- standing the experimental procedures and results
ports are clear, accurate, and concise, and contain must be in the main body of the text. Your demona comprehensive discussion of the results and their strator should not need to read the appendix.
Formal reports that are longer than the permitsignificance. They should be intelligible without
reference to the lab. script. You must not copy ted page limit or written in too small a font size
any sections directly from the script, text books or will be awarded a maximum mark of 12/20.
journal articles, but develop the skill of expressing
All reports should contain the following sections:
your understanding of the experiment in your own
words. By trying to express an idea, you will often
Title
deepen your understanding of it. If you do quote
Author
from another source, whether it is text or numeriAbstract
cal values, give a reference to the source including
Introduction
the pages where the quotation is located.
Theory
Experimental Method
The style of the report should be similar to that
Results
in a scientific journal. Take a look at journals in
Discussion
the library such as Physical Review B, Journal of
Conclusions
Applied Physics, and Proceedings of the Royal SoReferences
ciety. The technicians keep some good examples
of previous formal reports in the Level 4 Laboratory and you are encouraged to look at these before This order is recommended but not fixed. You may
starting on your first report.
feel, for reasons of style and clarity, that you wish to
All reports should be word processed and saved deviate from this precise format. You may also wish
in adobe pdf format (if you are unsure of how to to include Acknowledgements between the Concludo this, ask your demonstrator). Software is avail- sions and References.
able in room 1.14 if necessary. If you are unable
It is important to pay attention to literary style,
to produce a report electronically you must inform grammar, spelling etc. You will often find that in
your student and staff demonstrator immediately. an attempt to express precisely what you mean,
The upload time of your electronic copy is your ideas about the experiment and physics will
your official submission time.
be clarified. When you have finished, always read
32
4.5. FORMAL REPORT TEMPLATE

through carefully what you have written. Remember that written English is the vehicle by which
others are informed of what you have done and of
your ideas. The ability to communicate well is a
very important transferable skill.
4.3
Marking
All formal reports must be submitted electronically, via Blackboard, as well as on paper. The
submission time will be the upload time of the electronic copy (see Chapter 1 for details of penalties
for late submission). Further details will be available from your demonstrator.
A demonstrator will mark your formal report according to the mark scheme, and a staff demonstrator will examine the report and marking, and
moderate the mark where it is necessary to do so.
33
Never copy sections from lab. scripts. If you
have taken good notes you should not need to
refer to the scripts when writing your reports.
Never copy text verbatim either from another
student (even your lab. partner) or from a textbook or online source. Always re-write material in your own words.
Never cut and paste material from a website.
This includes figures and photographs.
Where you wish to quote material from another source, use quotation marks, and give
the source in your list of references. Quotation is appropriate, for example, for statements
of physical laws, but should definitely not be
used for whole paragraphs of text.
Avoid the use of Wikipedia scientific papers
or the recommended textbooks are generally a
more reliable source of information.
Formal reports are assessed only on their content

and not on the associated note taking and practical
You should be aware that plagiarism is considwork as these have already been assessed.
ered to be a very serious matter by the UniverThe mark scheme for the formal reports is shown sity and cannot be accepted. Further guidance on
in Chapter 5 and will be thoroughly discussed in avoiding plagiarism can be found at:
your lab. lectures in the Autumn term.
http://www.bris.ac.uk/is/library/
The second and third formal reports contribute
findinginformation/plagiarism
20% each to your final mark for the year. The
demonstrator will indicate clearly on your reports If you have questions concerning writing formal reareas in which improvements need to be made. He ports or plagiarism in general, please speak to your
or she will also complete a summary sheet indi- demonstrator or the staff demonstrator.
cating your performance in several areas and key
points that you need to note to improve your work.
Your first report will be marked in exactly the
same way as the later reports. You may find you
are awarded a relatively low mark with lots of comments and feedback. If you take good notice of this
advice you should be able to achieve a high standard in your later formally assessed reports. The
mark you are awarded for this report is for formative assessment only. However, if it is not of a
pass standard (i.e. you have not heeded the advice in this section) you will be required to rewrite,
re-submit and obtain a passmark for it.
4.4
Plagiarism
All formal reports should be submitted both as a

hard copy (to the first year laboratory) and electronically, via Blackboard. The electronic copies
will be checked for plagiarism by an automatic system that compares reports against those of other
students and against commonly available sources
on the internet. To avoid falling foul of the plagiarism checker, you should think about the following
points when writing your formal report:
4.5
Formal Report Template
When your report is marked it will be returned to

you with an orange feedback sheet. This sheet contains a number of check boxes to indicate the areas
in which your work is particularly strong or weak.
These categories are not used to directly calculate
your mark they are merely a list of common problem areas and the mark scheme is much more detailed but you should keep this sheet for reference
when you write your next report.
Your demonstrator will also add some general
feedback (a summary of the main points which will
be covered in more detail throughout your report)
and then a mark out of 20. This mark will be reviewed by the moderator, and in rare circumstances
there may be comments added, or an addition or
subtraction of overall marks in the Mark if moderated section. After the mark out of 20 is adjusted
for moderation, penalties for late submission will
be subtracted, and your final mark entered at the
bottom of the sheet.
The following page contains a template of a formal report which describes the appropriate information to be included in each section, and some
common mistakes.
34
CHAPTER 4. HOW TO WRITE FORMAL REPORTS
A Brief Formal Report Template

A. N. Author
This is the abstract. It does not require a section heading. The abstract should be a short summary
of the aim of the experiment, the basic method, main results and conclusions of your work. You should
present key numerical values (and their associated errors) in the abstract, and compare them to any
literature values. The abstract should not exceed 150 words and should not include any numbered
references. If a numbered reference is necessary to put your result in context, the details should be
included in the text of the abstract, not in the list of references at the end of the report.
1. Introduction
The Introduction sets the scene for your work. Here you explain the motivation for the work, and set it
in context.
Begin by explaining the importance of the work. Why do the measurements you have made matter?
Next, provide a succinct review of other attempts to make similar measurements and highlight any
shortcomings in this previous work. If you need to refer to work by other people, reference the original
sources like this [1]. Generally the introduction section will be one-third to three-quarters of a page long.
2. Theory
In this section you lay out the theoretical background to the measurements you will be reporting. You
should also include a brief summary of the key points of any relevant theories that you might be testing.
Any novel theoretical methods should be described in detail and any mathematical methods used in
the interpretation of the data clearly explained. You should not however give detailed derivations where
a suitable reference can be given.
Each equation that you include should be written on a separate line, and numbered on the right
hand side (as in equation 1) so that you can refer to them later. Use an equation editor to write your
equations. If you dont know how to do this ask a demonstrator. You should ensure that all variables in
your equations are defined in your text (not as bullet points), and are shown in a different font to the
main text (italicised times roman and greek characters are the most common), i.e., the energy, E, of a
photon in a vacuum is defined as
E=
hc
,
(1)
where h is Plancks constant, c is the speed of light in a vacuum and is the wavelength of the photon.
3. Experimental Method
In the Experimental section you should describe the equipment you use. If it is commercially available
equipment then provide the name and manufacturer. If it is home made then a full description is needed
sufficient for the reader to build their own.
You will often need to include diagrams illustrating the equipment used. These must be clear and
should have all the important components labelled. There must also be a suitable figure caption situated
below the figure. The text of the report should include a full description of the equipment which refers
back to the diagram (see Figure 4.1) so that the reader can see how it is all assembled, however a simple
list of apparatus should not be included.
Diagrams should not be overly elaborate and should be produced electronically or as a scanned high
quality diagram. Illustrations should not be copied from the Web or scanned in as this is
another form of plagiarism and may also breach copyright. (Note, in professional publications it
is possible to request permission to copy diagrams, suitably attributed, from the holder of the copyright,
but you shouldnt need to do this in your reports).
35
Speed controlled
motor and torque
sensor
Model flipper
Figure 1: A sketch of the constant flow apparatus used to measure the drag caused by
flipper bands.
In the experimental method section, you should describe the sequence of measurements you made
with comments about particularly difficult or unusual techniques used. You can think about this as a
cookery recipe the ingredients and the instructions must be sufficient to enable the reader to make the
dish. However, this must be written as good, flowing prose not as a series of bullet points or
list of instructions.
Note that taking care when taking scientific measurements is a self-evident requirement. It should not
be emphasised in a report.
4. Results
The Results section is where you actually report the measurements you made and the derived quantities
the data should usually be presented in graphs, occasionally tables. It is not normally necessary to
tabulate all your readings, but simply present a few typical data and the results derived from them. For
example if you measured the same quantity 10 times to get an idea of the accuracy of the measurement
you should only quote the mean value and its error. (In modern publications it is increasingly common
to supply data electronically as a supplement. You do not need to do this in your reports).
Your data should be analysed and the errors calculated. You do not need to give step-by-step details of
the error calculations but you should state how the procedure was carried out, with a single final equation
where necessary and references where appropriate. It is not necessary to include standard statistical
formulae (e.g. standard deviation, or standard error in the mean). Error analysis is a compulsory part
of the results.
It is very important to remember that the results section should be in prose style referring to tables
and graphs as necessary. It is not acceptable simply to say the results are shown in figures 1 to 4 and
not describe them.
Displaying your results
If your data can be presented meaningfully in a graph or a table, then a graph should be chosen. It
is both quicker and easier to see trends when data is presented in a graph. If your data is presented
graphically, then there is generally no need to include it in a table as well.
When using graphs to present results you should ensure the axes are clearly labelled and the units
included. The scientific graphing package Origin is available to all science faculty students and should
be used in preference to Excel where possible. Origin allows you to use scientific notation, symbols and
super- or sub-scripts when writing legends and axes so that professional graphs can be produced easily.
Graphs should also include error bars indicating the reliability of the results. Unless there is a good
theoretical reason, you should not draw in lines joining up the measured data points. If a slope is
calculated the line should be clearly visible on the graph, although normally the equation of the line
would be removed and mentioned in the text.
When producing graphs take care to select appropriate axes and colours. Graphs produced directly by
e.g. Excel generally need to be edited to a suitable style (for example computer style scientific notation
36
CHAPTER 4. HOW TO WRITE FORMAL REPORTS
numbers should not appear on your graphs). The scientific graphing package Origin is available to all
science faculty students, and should be used in preference to Excel where possible.
You must always include a caption to inform the reader what the figure shows (see, for example,
Figure 2). Graphs do not normally need a title - a figure caption below the graph should be sufficient.
1.9
1.8
Relative power
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1
0
0.5
1.5
2.5
3.5
4.5
Speed (m/s)
Figure 2: A graph showing the relative power that has to be supplied to the test apparatus to
maintain a constant speed when a conventional flipper band is attached to the model flipper.
The square symbols relate to the porpoising experiments and the diamond shaped symbols
to the constant flow experiments.
Note that in scientific literature diagrams and graphs are referred to as Figures and should be
numbered in the order they appear in the text. Tables are referred to as Tables and should have
separate numbering. The captions for tables are normally positioned above the table, while figure
captions should be below the figures.
5. Discussion
The discussion is where you look at the results you have obtained and place them in their context. You
should discuss the theoretical grounds for expecting the results obtained, and whether or not your results
fit your expectations. You should note the errors in your measurements and assess the reliability of your
data compared to that of others. It is important to show careful and critical judgement in this section.
If your data are not consistent with theory or previous work, you will need to discuss the reasons and
also to come to a judgement as to whether your work is better and if so why. You should also consider
what changes you could make to the apparatus or methods to improve the results, and suggest methods
to minimise each type of experimental error.
6. Conclusion
Here you simply restate the main conclusion of your work, usually in a single paragraph. This might be
a simple quantitative statement such as:
the maximum swimming speed of a King penguin is 12 1 m/s
or it might be qualitative such as:
The physiological differences between Macaroni penguins (Eudyptes chrysolophus) and Royal
penguins (Eudyptes schlegeli) are sufficient to justify their classification as separate species.
In general there will be both qualitative and quantitative conclusions to be drawn. For example:
We have demonstrated the ability of Emperor penguins (Aptenodytes forsteri) to forage for
prey over most of the Southern Oceans south of the Antarctic convergence. Further detailed
diving studies reveal the ability of these birds to dive to depths in excess of 500 m and to remain
submerged for more than 18 minutes.
Dont worry if your conclusions sound very like your abstract they should.
37
7. References
This is an extremely important section. You should ensure all references are accurate since they may be
checked.
References should all be numbered consecutively in the order they appear in the text. Each reference
should appear only once, regardless of the number of times it is mentioned in the text.
If you are referencing a paper or article ([1], [2]), you should list the author(s), journal, date, volume
and page number.
If you are referencing a book ([3]), you should give author(s)/editor(s), title, edition, publisher and
where appropriate chapter and page number.
For example, in this case the list of references is,
[1] A.D. Physicist, Some rather old paper written a long time ago, Phys. Rev. D, 386, 4 6 (1901).
[2] I.M.N. Author, This is what I wrote, Journal Name, Volume No., page range (year).
[3] P.A. Tipler and G. Mosca, Physics for Scientists and Engineers, 5th Edition, W.H. Freeman and
Company, New York (2004), Chapter 15.
Formats vary (and in reality must satisfy the style of the journal to which you are submitting), but
be sure to include the information shown so that your readers can find the references for themselves, if
they need to.
For the purposes of first year reports you should not include references to Web pages. In general, if the
work is suitable for referencing, the Web page will give a reference to a formally published reference. For
example a good Wikipedia article will have formal references, that you should read and include in your
own list of references. Simple references to Web pages with the date of access are unacceptable. The
reason for this is that most web pages are liable to be changed (updated), deleted, or lost (as servers are
shutdown) and there is no global and formal method of archiving them. There are methods for properly
referencing journal articles electronically but these will be introduced to you in later years.
Examples of incomplete or otherwise unacceptable references are:
Physical Review Letters
A.N.Other Physical Review Letters (2008)
Classical Physics
http://www.wikipedia.org
http://en.wikipedia.org/wiki/Newton
http://www.phy.bris.ac.uk/groups/particle/research.html
http:// www.wikipedia.org/wiki/Newton accessed 17/7/2009
Chapter 5
Mark Schemes
Chapter 1 explained that three types of assign- methods, and will show evidence of extended readment contribute to your final laboratory grade: pre- ing around the subject, well beyond suggested furlabs, laboratory notebooks and formal reports.
ther reading.
The pre-lab tests have well-defined answers and Aims: Understanding of the overall aims of the
are marked out of 100. Late submissions are subexperiment.
ject to late penalties (section 1.11) and prevent you
Background reading: Evidence of extended reafrom starting experiments on time.
ding and an ability to relate it to the experiLaboratory books and formal reports are both
ment.
marked out of 20 the mark schemes are detailed
here. The marks can be converted to a percentage Understanding of theory
to work out your final grade, but the conversion
scale is non-linear (i.e., 8/20 6= 40 %). Assignments Understanding of experimental methods
marked out of 20 will accrue late penalties of 3
marks per day (section 1.11).
Experimental Results
This section assesses the experimental skill and dili5.1 Practical Work Marking gence needed to obtain reliable results. Allowance
will be made here for major problems encountered
Criteria
due to equipment malfunction etc. and whether
the student was working alone or in a pair.
Your marked lab books will be returned to you with
Excellent students will produce extensive high
handwritten comments throughout and an orange quality data with evidence of hard work and exFeedback Form attached. The form has a box perimental skill.
for the marker to summarize your main strengths
and weaker areas. It also contains a series of check Quality and Quantity of Data Obtained: In
boxes where your performance in various categories
cases where the experiment has not worked as
is graded to further help you identify areas for imexpected there should be clear evidence that
provement. The assessment mark for the work canall reasonable steps have been made to get
not be determined from these check boxes using a
the results.
numerical algorithm. The four assessment criteria
used to calculate your mark are: Understanding,
Experimental Results, Notebook & Organ- Notebook and Organisation
isation and Critical Sense & Creativity. An
outline of the categories considered when assessing This section assesses the records kept throughout
the experiment, including notes on apparatus, thethese four criteria is given below.
ory and data obtained. All notes should be kept
in a bound notebook, with graphs, tables etc. securely attached.
Understanding
An excellent notebook will contain a concise and
This section assesses the understanding of the theo- well thought out summary of what is planned for
retical background and the goals of the experiment. each lab day and (at the end) what was achieved
It also assesses the understanding of the function of in each day. It will be very well organised, with exthe experimental apparatus and the data analysis tensive graphs and analysis embedded in the notes
techniques.
at appropriate points. This also applies to experiExcellent students will demonstrate a thorough ments which are either based on or make extensive
understanding of the theory and the experimental use of computers.
38
5.2. FORMAL REPORT MARKING CRITERIA
39
Clarity and Organisation: The notes should be Autonomy: Assesses the degree to which the stueasy to follow, mistakes should be neatly
dent worked independently (after suitable incrossed out (do not tear out pages) with a brief
duction), without requiring large amounts of
explanation as to what was wrong. Summaries
oversight from the supervisor.
should be written at appropriate points.
Improvements / Extensions: Sensible and creative / original extensions or refinements to
Theory and Experimental Procedure: Some
the experiment have been implemented and as(but not extensive) notes on theory and
sessed, after consideration of the limitations of
experimental procedures included. A concise
the equipment and sources of both random and
description of all relevant parts of the expersystematic errors.
iment and a description of analysis methods
used.
Data collection: Experimental data should be in
neat tabular form (where appropriate) with
units, uncertainties and all relevant experimental conditions/settings noted. In cases
where data is collected by computer it is still
important to note relevant conditions and also
the names of the files in which the data is
stored.
Graphs & Diagrams: Should be neat, labelled,
and embedded in the notes at appropriate
points (not stapled in en masse at the end of
the experiment.)
Critical Sense and Creativity

This section assesses the critical and mathematical analysis of the data and results, and whether
appropriate procedures have been devised to overcome any limitations of equipment based on the
results obtained. Account will be taken of the degree to which students have been able to work independently of their supervisor, and have devised
experiments and procedures independently.
5.2
Formal Report Marking

Criteria
Your marked reports will be returned to you with

handwritten comments throughout and an orange
Feedback Form attached. The form has a box for
the marker to summarize your main strengths and
weaker areas. The form also contains a series of
check boxes where your performance in various categories will be graded to further help you to identify
areas for improvement. The assessment mark for
the report cannot be determined from these check
box criteria using a numerical algorithm. The four
assessment criteria used to calculate your mark are:
Clarity & Fluency, Scientific Content, Style
& Presentation and Criticism & Insight. An
outline of the categories considered when assessing
these criteria is given below.
Clarity and Fluency
This section assesses the quality of writing, the flow

of the report and how well the explanations and
arguments are made.
An excellent report will be easy and engaging
Excellent students will have an extensive and
to read, with the material laid out in a coherent,
thorough data analysis throughout the experiment,
logical and accessible manner.
with ongoing error analysis. The analysis and discussion (of the results and methods) will include
Coherence: The narrative should flow logically
references to literature perhaps resulting in the use
and easily from one section to another. There
of novel techniques.
should be good links between the figures and
the text.
Data analysis: Assesses the depth of analysis and
the degree to which it is free from error.
Concise: All information should be relevant and
necessary. Unnecessary material, for examError analysis: Uncertainties in measurements
ple trivial or standard theoretical derivations,
should be considered throughout. Graphs
should not be included.
should generally include appropriate error
bars, and a mathematical error analysis is re- Explanations: The objectives, theory, methods
and data analysis techniques should be easy
quired where appropriate.
to follow, not simply a series of unconnected
Planning and Continuous Assessment &
points.
Development:
Evidence of planning
throughout the experiment, with the ability Arguments: The discussion and interpretation
should be clearly expressed in a logical and
to refine procedures after consideration of new
thoughtful manner.
data or information.
40
CHAPTER 5. MARK SCHEMES
Scientific Content
of graphs, diagrams and tables. It also assesses

grammar, spelling and the general literary style of
This section assesses the scientific quality of the re- the document.
port; how well it demonstrates understanding and
An excellent report will have a professional aprelevant concepts, the quality of the analysis, the
pearance
throughout with no errors in grammar or
depth of conclusions and the ability to compare to
spelling.
The
size of the symbols and lettering used
other work.
on all parts of a graph or diagram should be legiAn excellent report will include an original exble (this includes subscripts) as a guide a minimum
planation of background/theory, with evidence of
size of 2mm is suggested. Lines should have suffiextensive reading and understanding of relevant litcient weight (e.g. 0.5pt or 0.2mm). Normally titles
erature. The data analysis will include a thorough
for figures are not needed rather the relevant inerror analysis, and there will be extensive and deformation is included in the caption.
tailed conclusions drawn from the data, with references to prior results in the literature where approFormatting: Should be consistent throughout
priate.
(including use of bold or italics for labelling
etc.). Figure captions and equations should
Abstract: Should be self contained and concise,
be in the same font size as the text.
with a clear statement of the results and conclusions.
Layout: The report should be well laid out with
Background & Theory: The relevant backsections and subsections where appropriate.
ground/theory should be identified and corContent should be presented in the approprirectly explained. The key theoretical results
ate sections, and figures located in suitable popertinent to the data analysis should be desitions throughout the report.
rived or expressed, and the origins of relevant
Language: The report should be written in a proequations should be explained.
fessional scientific style, with the past, present
Apparatus & Methods: All relevant experiand future tenses used appropriately. The pasmental procedures and experimental apparasive voice should normally be used, although
tus should be described.
first person plural may be used where appropriate.
Data Analysis: Should be accurate and checkable by the reader where possible. Results
Spelling and Grammar: There should be no
should be expressed in appropriate units (ususpelling or grammatical mistakes.
ally S.I. with suitable prefixes e.g. k, M).
Error Analysis: Errors should be evaluated Equations: Should be clear with appropriate
symbols, numbered sequentially and correctly
properly, quoted in the correct places and to
cited in the text. All variables should be dethe correct precision. Sources of systematic
fined in the text (except standard symbols such
error should be noted.
as e, h, etc) and should be in italic script.
Discussion & Conclusions: Good scientific arguments should be made when interpreting the Diagrams: Should be used to illustrate concepts
quality and validity of results, and relevant litwhich are difficult to explain with text alone
erature should be used to support arguments
(e.g., the layout of the experimental apparaand conclusions. Comparisons to other work
tus). They should be clear, uncluttered and
should be cited properly.
well labelled with a suitable caption. Original
work is preferable as this may help to show
Referencing: Appropriate and relevant referoriginality. If figures are copied the source
ences should be identified, with excellent reshould be clearly identified (note this includes
ports showing evidence of reading beyond the
your partners work).
scope of that identified in the script. References should back up statements made in the Graphs: Should be clear and correctly labelled,
text rather than just be a list of things you
with appropriate axes, error bars, and font
have read.
sizes. Suitable captions should be included.
Tables: Should be clear with appropriate labels,
units and a suitable caption. Data should not
This section concerns the general layout and organbe presented in a table if it could be expressed
isation of the report, the quality of the presentation
better in a graph.
Style and Presentation
5.3. EXPLAINING THE 21-POINT MARKING SCALE

Internal referencing: Figures, equations and tables should all be referenced correctly and consistently within the text. Any items not discussed in the text should not be included (e.g.,
graphs of data which are not described in the
text).
Referencing style: One of the standard referencing methods should be used consistently (Numerical [1] or Harvard (Smith 2010)).
Criticism and Insight

This section assesses your ability to objectively criticise your own and others work, suggest improvements or extensions (were appropriate) and portraying your work in a wider context.
An excellent report should demonstrate originality and a thorough understanding of the scientific
principles involved.
Critique of methods: Discussion of the limitations of the experiment and sensible suggestions for improvements to the experiment
should be given were appropriate.
Evaluation of results: A careful discussion of
the results should be presented, including the
relevance and significance the errors. A thorough comparison with other techniques or previous measurements should be made.
Creativity: The experimental results and/or
wider reading should be used to generate novel
suggestions or explanations where appropriate.
Originality: There should be evidence of independent thought and an effort to explain concepts
in an original manner.
Wider context: The wider relevance of the results and how they may affect or be important
for other fields of work should be discussed.
5.3
Explaining the 21-point

marking scale
The check-boxes on the orange feedback sheets

show your achievement in each assessment category, and these guide the award of a single mark
out of 20 according to the table on the following
page. Further details are available on the DLM.
As previously stated, your marks on the 21-point
scale do not convert linearly to a percentage. The
following table shows the conversion between the
two scales.
41
Mark out of 20
Equivalent percentage
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
100
94
89
83
78
72
68
65
62
58
55
52
48
45
42
35
29
22
15
7
0
42
CHAPTER 5. MARK SCHEMES
The 21-point Marking Scheme

The performance in each of the assessment criteria is used to award a mark according to the table below.
Mark
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
4
3
2
1
Criteria to be satisfied
Work is close to publishable standard/worthy of dissemination.
Mastery of advanced methods and techniques at a level beyond that explicitly taught.
Ability to synthesise and employ in an original way ideas from across the subject.
Excellent presentation.
Outstanding command of critical analysis and judgement.
Excellent range and depth of attainment of intended learning outcomes.
Mastery of a wide range of methods and techniques.
Evidence of study and originality clearly beyond the bounds of what has been taught.
Excellent presentation.
Able to display a command of critical analysis and judgement.
Attained all the intended learning outcomes for this unit.
Able to use well a range of methods and techniques to come to conclusions.
Evidence of study, comprehension, and synthesis beyond the bounds of what has been
explicitly taught.
Very good presentation of material.
Able to employ critical analysis and judgement.
Some limitations in attainment of learning objectives, but has managed to grasp most
of them.
Able to use most of the methods and techniques taught.
Evidence of study and comprehension of what has been taught.
Adequate presentation of material.
Some grasp of issues and concepts underlying the techniques and material taught.
Limited attainment of intended learning outcomes.
Able to use a proportion of the basis methods and techniques taught.
Evidence of study and comprehension of what has been taught, but grasp insecure.
Poorly presented.
Some grasp of the issues and concepts underlying the techniques and material taught,
but weak and incomplete.
Attainment of only a minority of the learning outcomes.
Able to demonstrate a clear but limited use of some of the basic methods and
techniques taught.
Weak and incomplete grasp of what has been taught.
Deficient understanding of the issues and concepts underlying the techniques and
material taught.
Attainment of nearly all the intended learning outcomes deficient.
Lack of ability to use at all or the right methods and techniques taught.
Inadequate and incoherently presented.
Wholly deficient grasp of what has been taught.
Lack of understanding of the issues and concepts underlying the techniques and
material taught.
No significant assessable material, absent, or assessment missing a must pass
component.
Part II
The Experiments
43
Experiment C
Measurement of the speed of light

Aims:
1. To measure c, the speed of light, by three different techniques.
Objectives:
1. To gain experience of precision measurement techniques.
2. To gain further experience with the analysis and elimination of experimental errors.
3. To gain an understanding of the methods used to align equipment.
4. To gain familiarity with modern optical apparatus.
Microscope
Matt
Screen
Laser
Attenuator
L1 ,
f = 50mm
Beam
Splitter
Digital
Micrometer
L2 ,
f = 250mm
MR Speed
Control
MR
Figure 1: The Foucault apparatus for measuring the speed of light.

Further reading:
P.A. Tipler & G. Mosca, Physics for Scientists and Engineers 5th Edn. (W.H. Freeman and
Company, New York, 2004). Chapters 31 and 32.
45
46
EXPERIMENT C.
MEASUREMENT OF THE SPEED OF LIGHT
Introduction
The velocity of light in free space is one of the most
important and intriguing physical constants. Many
methods of measuring it have been devised over the
centuries. At the end of the 19th century the careful measurements of Michelson and Morley demonstrated that it was, perhaps counter-intuitively,
constant for all inertial reference frames. This result led to the fundamental postulate of Einstein
that the speed of light in a vacuum c is constant
and independent of any uniform (non accelerating)
motion of the observer. This led directly to Einsteins Special Theory of Relativity.
Whether the light comes from a laser on a desktop or from a star hurtling away at fantastic speed,
if you measure the velocity of the light in a laboratory, you measure the same constant value. In more
precise terminology, the velocity of light is independent of the relative velocities of the light source and
the observer. In particular theory shows that:
MR
(rotating
mirror)
Quarter-wave
plate
s
L2
PM
Beam-splitter
MF
(fixed mirror)
L1
Laser
PL
s'
Measuring
Microscope
Figure C.2: Diagram of the Foucault method
Consider a beam of light leaving the laser. First

the beam is focused by L1 to a point at s, then
it passes through the beam-splitter and L2 and is
reflected from the rotating mirror, MR . It is then
focused to a point on the fixed mirror MF , which
reflects the light back along the same path to MR .
The beam then returns through the beam-splitter,
and is focused to a point at position s0 , where it can
(1) the velocity of light establishes an upper limit
be viewed through the measuring microscope. This
to the velocity that may be imparted to any
beam of light is reflected from a particular point on
object,
MF which is determined by the angle of MR .
(2) objects moving near the velocity of light follow
a set of physical laws drastically different, not
only from Newtons Laws, but from the basic
assumptions of human intuition.
A result of the fundamental nature of the velocity

of light was such that in 1983 the S.I. definition of
the metre was recast in terms of the fundamental
definition of the second and by defining the velocity
of light as exactly 299,792,458 m.s1 .
The title of this experiment may now seem rather
meaningless - we cannot measure the speed of light
as it is a defined fundamental constant! In a modern context you may think of these experiments
as methods of calibrating your length measuring
apparatus. The experiments reflect the historical
(pre-1983) methods used to determine the speed of
light. You should look into the methods of Galileo,
Romer, Fizeau and especially Foucault.
C1: The Foucault method

The experiment you will be carrying out is a modern version of the technique established by Jean
Foucault. The experimental arrangement for the
Foucault method is shown in Figure 2. PL and
PM are polarisers, and L1 and L2 are lenses. The
quarter-wave plate can be moved in and out of the
beam path. When the components are aligned correctly, the polarisers and quarter-wave plate do not
affect the beam paths in the system.
Figure C.3: The reflection point on MR . (a) When

MR is at angle , the laser beam is reflected to point
S on MF . (b) When MR is at angle 1 = + ,
the laser beam is reflected to point S1 on MF .
As the first step in the derivation, we must determine how the point of reflection on MF relates
to the rotational angle of MR . Figure C.3(a) shows
the path of the beam of light, from the laser to MF ,
47
when MR is at an angle . In this case, the angle
of incidence of the light path as it strikes MR is
also and, since the angle of incidence equals the
angle of reflection, the angle between the incident
and reflected rays is just 2. As shown in the diagram, the pulse of light strikes MF at a point that
we have labelled S.
Figure C.3(b) shows the path of the pulse of light
if it leaves the laser at a slightly later time, when
MR is at an angle 1 = + . The angle of incidence is now equal to 1 = + , so that the angle between the incident and reflected rays is just
21 = 2( +). This time we label the point where
the pulse strikes MF as S1 . If we define D as the
distance between MF and MR , then the distance
between S and S1 can be calculated:
S1 S = D(21 2) = D[2( + ) 2] = 2D
(C.1)
In the next step, think of a single short pulse of
light leaving the laser. Suppose MR is rotating, and
this pulse of light strikes MR when it is at angle
(Figure C.3(a)). The pulse will then be reflected
to point S on MF . However, by the time the pulse
returns to MR , MR will have rotated to a new angle,
say angle 1 . If MR had not been rotating, but had
remained stationary, this returning pulse of light
would be refocused at point s. Clearly, since MR
is now in a different position, the light pulse will
be refocused at a different point. We must now
determine where that new point will be.
The situation is very much like that shown in
Figure C.3(b), with one important difference: the
beam of light that is returning to MR is coming
from point S on MF , instead of from point S1 . To
make the situation simpler, it is convenient to remove the confusion of the rotating mirror and the
beam splitter by looking at the virtual images of
the beam path, as shown in Figure C.4.
virtual images, the problem becomes a simple application of thin lens optics (see Level 4 Oscillations
& Waves lectures, and Tipler & Mosca). With MR
at angle 1 , point S1 is on the focal axis of lens L2 .
Point S is in the focal plane of lens L2 , but it is
a distance S = S1 S away from the focal axis.
From thin lens theory, we know that an object of
height S in the focal plane of L2 will be focused
in the plane of point s with a height of (v/u)S.
Here v and u are the distances of the lens from
the image and object, respectively, and the minus
sign corresponds to the inversion of the image. As
shown in Figure C.4, reflection from the beam splitter forms a similar image of the same height.
Therefore, ignoring the minus sign since we
arent concerned that the image is inverted, we can
write an expression for the displacement (s0 ) of
the image point:
s0 = s =
A
v
S =
S
u
B+D
(C.2)
Combining equations (C.1) and (C.2), and noting

that S = S1 S, the displacement of the image
point relates to the initial and secondary positions
of MR by the formula:
s0 =
2DA
B+D
(C.3)
The angle depends on the rotational velocity of

MR and on the time it takes the light pulse to travel
back and forth between MR and MF , a distance of
2D. The equation for this relationship is:
=
2D
c
(C.4)
where c is the speed of light and is the rotational

velocity of the mirror in radians per second. (2D/c
is the time it takes the light pulse to travel from
MR to MF and back).
Using (C.4) to replace in (C.3) gives:
s0 =
4AD2
c(B + D)
(C.5)
Equation (C.5) can be rearranged to provide our

final equation for the speed of light:
c=
4AD2
(B + D)s0
(C.6)
where:
Figure C.4: Analyzing the virtual images.
c = the speed of light;

= the rotational velocity of MR ;
The critical geometry of the virtual images is the

same as for the reflected images. Looking at the
A = the distance between lens L2 and lens L1 ,

minus the focal length of L1 ;
48
EXPERIMENT C.
B = the distance between lens L2 and MR ;

D = the distance between MR and MF ;
s0 = the displacement of the image point,
as viewed through the measuring microscope.
(s0 = s1 s; where s is the position of the image point when the rotating mirror is stationary, and s1 is the position of the image point
when the rotating mirror is rotating with angular velocity .)
Equation (C.6) was derived on the assumption
that the image point is the result of a single, short
pulse of light from the laser. But, looking back
at equations (C.1)(C.4), the displacement of the
image point depends only on the difference in the
angular position of MR in the time it takes for the
light to travel between the mirrors. The displacement does not depend on the specific mirror angles
for any given pulse.
If we think of the continuous laser beam as a
series of infinitely small pulses, the image due to
each pulse will be displaced by the same amount.
All these images displaced by the same amount will,
of course, result in a single image. By measuring
the displacement of this image, the rate of rotation
of MR , and the relevant distances between components, the speed of light can be measured.
Method
Safety Notice:
This experiment involves the use of a 1 mW
laser which has the potential to cause eye
damage if you stare directly into the beam.
When the mirror is stationary you should only
observe the beam indirectly on a piece of
paper, or through crossed polarisers. Do not
look into the unattenuated laser beam, either
directly or as it reflects from a mirror. Ask
your demonstrator for advice if you are unsure
about any safety aspects of the experiment.
To align the equipment:

Refer to Figure C.1 and Figure C.2.
(a) Unblock the laser beam and use your fingers to
rotate the drive wheel of the rotating mirror,
MR such that the laser beam is reflected down
the lab towards the fixed wall mirror, MF .

(b) Adjust the fixed wall mirror, MF , using the
three screws, such that the incident beam is
returned along the same path to the rotating
mirror. This can be a difficult process. There
are two ways to achieve this.
Firstly, by eye - look down the lab to the rotating mirror and with patience, adjust the fixed
mirror until the beam can be seen on the white
surround of the rotating mirror.
Secondly, tear off a small piece of paper. Move
the paper in the path of the reflected beam until you can see it on the torn edge. It is then
possible to walk towards the rotating mirror
using the paper as a guide to the path of the
reflected beam. This path can be adjusted using the screws on the fixed wall mirror until it
strikes the rotating mirror. What advantage is
there to using a torn edge to locate the beam?
This step is easiest with two people: one adjusting MF , and one watching the beam position on MR .
(c) Use the three alignment screws on MF to reflect the incident beam directly back to the
centre of MR . Position the tracing-paper matt
screen (on the free standing base) between the
beam splitter and the microscope. The reflected beam can be clearly seen as a point
of light. To verify that this point is the reflection, block the beam travelling to MF with
your hand. The point should extinguish. N.B.
You should not attempt to realign any other
part of the apparatus. If you think there is a
problem with the alignment, please discuss the
matter with your demonstrator.
(d) With the beam attenuated (polarisers
crossed), use the microscope to locate and focus the reflected beam image. Adjust the microscope to bring the reflected beam image
into the centre of the cross-hairs.
Making the measurement:

The speed of light measurement is made by rotating
the mirror at high speeds and using the microscope
and micrometer to measure the corresponding deflection of the image point.
(1) With the apparatus aligned and the beam
image in focus (see previous section), take
a measurement of the position of the microscope. You should use both the Vernier scale
on the microscope and the digital micrometer
attached at the side. When using the digital
micrometer, make sure that you have 2 mm of
travel in both directions.
49
Important The Rotating Mirror:

The rotating mirror assembly is delicate and
must be used carefully. Before turning on the
motor, make sure the mirror rotates freely by
hand. Whenever the motor is accelerated, the
red light on the front panel of the motor control box will light up. As the speed stabilizes,
this light should go off. If it does not, turn off
the motor - something is interfering with the
motor rotation. Check the drive mechanism
is free before continuing.
Never run the motor with the MAX
REV/SEC button pushed for more than one
minute at a time and always allow one minute
between runs for the motor to cool.
(1) Set the direction switch on the rotating mirror power supply to CW and turn on the motor. Through the microscope, you should see
a small, very faint spot that moves as the rotating mirror changes speed. Verify that you
are looking at the correct point by increasing
and decreasing the speed of the rotating mirror, and blocking and unblocking the beam to
MF so that the spot flashes. You can vary
the intensity by adjusting the microscope polariser, PM . For your safety, take care to
ensure that the mirror does not come to
a complete stop whilst you are looking
through the microscope.
It is easy to confuse the correct spot with other
brighter spots caused by reflecting surfaces in
the beam path. In practice, you will not obtain a pin-sharp beam image when the mirror
is rotating; the beam image will be a slightly
diffuse spot or a narrow bar. In the latter case,
it is recommended that you repeat the alignment process with a different portion of the
fixed wall mirror (raise or lower it). If this
fails to work, then you must make your best
estimate of the centre of the bar and use this
in your measurements. (You may be able to
improve the beam image by adjusting the microscope, after which you must stop the mirror
and repeat your zero point reading.)
(2) Let the motor warm up for 3 minutes at about
600 revolutions per second. Record the rotational frequency at ten second intervals for a
couple of minutes and observe the variation
in the rotational motor speed, . You should
observe random fluctuations. If you find a
systematic trend, the motor has not settled
and you will need to repeat the measurements.
This will give you a measure of the error in motor speed, . N.B. if you find large fluctuations in i.e. greater than the least significant
digit of the display, the readout may be faulty,
and you should notify your demonstrator.
(3) Now you can establish the error in a single
reading of s0 . To do this, keep the mirror
rotating at a constant speed and repeatedly
bring the cross-hairs onto the beam image and
record the reading on both the microscope and
the digital micrometer. You should take about
ten independent readings, from which you can
establish the error, s0 .
(4) Record the beam deflection for rotational
speeds between 100 and 1000 rev/sec. Think
about the accuracy of your measurements and
if necessary, repeat your readings.
(5) For one final reading, use the ADJUST knob
to bring the rotational speed up to about 1,000
rev/sec. Then push the MAX REV/SEC
button and hold it down. When the rotation
speed stabilizes:
move the travelling microscope to bring
the beam image into the centre of the
cross-hairs;
record the rotational speed and release
the MAX REV/SEC button;
record the position of the travelling microscope, using both the Vernier scale and
the digital micrometer.
When the mirror rotates at 1000 revs/sec or
more, the image may widen slightly in the direction
of displacement. Position the microscope cross-hair
in the centre of the resulting image.
The optical components used in this set-up are
reasonably simple - even the polarisers are used
solely to reduce the light intensity to a comfortable
level. In the next section, you will be introduced to
more complicated optical components which make
use of the polarized state of the laser to isolate a
portion of the beam. You will repeat the experiment using these components, and this should allow you to assess whether they improve the data.
Improving the data collection with

modern optical techniques
You have obtained data with the mirror rotating
in one direction (either clockwise, CW, or counterclockwise, CCW), but you will probably have observed a red background in the microscope, possibly
with extra laser spots which do not deflect as the
mirror speed is varied.
50
EXPERIMENT C.
These extra spots and background light are reflections from various surfaces of the other optical components in the system. (If you block first
MR then L2 you should see some spots disappear.)
They are an indicator of how well-aligned the system is - a perfect alignment would result in one single spot coincident with the spot from the rest position of MR . Clearly this perfect alignment would
make the desired spot impossible to detect so a
slight mis-alignment is actually necessary. However, this extra light does mean that when MR is
rotating the correct spot can be hard to discern.
By using more sophisticated optical components
the unwanted light can be eliminated from the microscope. You will set up these components and
then obtain the displacement vs. rotational speed
data in the same manner as before but with the
mirror rotating in the other direction. Using this
new technique you can assess whether the new components improve either the accuracy of the data or
the ease of the data collection.
The potential improvement to the kit is outlined
below. Optics textbooks should provide more information if you wish to improve your understanding.

Now repeat your estimate of the error in the microscope reading (the error in the motor speed can
be assumed to be the same), then obtain values of
the displacement vs. motor speed as before.
Finally, take measurements of the distances A, B
and D, defined earlier. With these measurements
you can use Equation C.6 to determine c and its
error according to the methods you have considered in the pre-laboratory exercises. Remember to
take into account the fact that the rotational speed
given on the motor must be converted to an angular
velocity for the calculation.
Calculate c and its associated error both with
the quarter-wave plate and without, then discuss
the results.
C2: The Fibre-optic method

This experiment aims to measure the time-of-flight
of optical photons propagating along a fibre-optic
cable. The measurements are straightforward, with
all timings being measured with an oscilloscope.
The time of propagation may be given by:
nx
(C.7)
t = t0 +
The system already includes a polariser, PL ,
c
directly in front of the laser. This ensures that
where t0 is an unknown propagation delay caused
the light from the laser is linearly polarised.
by the electronics, both in generating and detecting
This polariser should not be rotated.
the light, n is the refractive index of the fibre-optic
A quarter-wave plate is fixed to the bench on and x its length. By performing measurements on
a hinged support between MR and MF , and cables of various lengths, t0 may be eliminated, and
should be carefully moved upright into the c determined from a graph of t versus x.
path of the beam. The wave plate is tuned
to the laser light at 633nm, and positioned so (1) Connect the short fibre-optic link (nominal
5m) between the light-emitting diode (LED)
that the the light leaving PL is aligned at 45o
transmitter and the fibre-optic receiver, as
to both the slow and fast axes of the wave
shown in the photograph.
plate. This has the effect of converting the
light to a circularly-polarised state.
(2) Connect the fibre-optic assembly to the power
When the light is reflected from MF it experiences a shift in handedness. As a result, when
it passes back through the quarter-wave plate
it is converted to linearly polarised light that
is 90o out of phase with the original light from
PL .
The polariser in front of the microscope, PM ,
can now be rotated to eliminate most background light, whilst allowing the spot that
travels to MF to remain unaffected.
Set up your system with MR stationary, so that
the spot travelling to MF is the only one visible in
the microscope, then turn on the motor so that it
is rotating in the opposite direction to previously.
You may need to rotate the microscope polariser,
PM , slightly to improve the image.
supply and switch on.

(3) Connect the TTL output of a signal generator
to the input of the LED transmitter.
(4) Connect the signal generator to channel 1 of
the oscilloscope and the output of the photodiode circuit to channel 2.
(5) Make sure that you can see two square-wave
traces on the screen (Figure C.6).
(6) If necessary, adjust the positions of each end of
the fibre-optic cable for maximum amplitude
of the detected signal.
(7) Expand the oscilloscope trace so that you can
see the time difference between the transmitted and detected signals. Measure the time
difference between the two signals, using the
51
Fibre optic
cables
LED
transmitter
LED Trx
power supply
Photodiode
receiver
Figure C.5: The experimental set-up for the fibre-optic method.
Ch. 1
Ch. 2
Formal Report Guidance:

If you choose to write about Experiment C for
one of your formal reports, you must discuss
at least two of the three methods below:
Foucault method without quarter wave
plate,
Figure C.6: Typical oscilloscope trace.
CURSOR controls on the oscilloscope, or

otherwise. You may find it helps to AVERAGE over multiple sweeps of the oscilloscope
to reduce the noise level of the detected signal.
You will have to decide which part of the trace
to take your measurement from.
(8) How accurate and reproducible are your measurements? Should you repeat them?
Note: The length of each fibre optic cable is

written on a marker on the cable. The error
in the length can be taken as 5mm per metre.
You should now be able to make your determination of c and its error according to the methods
you have considered in the pre-lab. exercise.
Foucault method with quarter wave

plate,
Fibre-optic method.
Experiment F
Optical Interference
Aims:
1. To use the phenomenon of interference to measure the radius of curvature of a spherical
lens and determine the refractive index of a liquid.
2. To produce a hologram and investigate the real and virtual images it produces.
Objectives:
1. To understand the concept of interference and gain an awareness of its application to measuring small distances.
2. To learn to use a travelling vernier microscope.
3. To encourage experimentation when producing holograms.
Figure F.1: The Newtons rings apparatus.
Further reading:
Halliday, Resnick & Krane, Physics, Volume 2, pp 958-957; 997-998.
F.G. Smith and J.H. Thompson, Optics, ch.13, Wiley (Manchester Series).
Company, New York, 2004). Chapters 33-2 & 34-4.
M.A. Morrison Understanding Quantum Physics (Prentice Hall).
A. Hey & P. Walters The Quantum Universe (CUP).
52
53
Introduction
(a) the quality of a glass lens
The properties of waves and the phenomenon of interference play an important part in many aspects
of physics ranging from the properties of sound, the
quantum nature of fundamental particles and the
behaviour of electrons in crystals to name but a few.
In this experiment you will investigate and quantify these effects in a classic experiment known as
Newtons rings. Details about this experiment can
be found in any good physics textbook on optics.
An important factor necessary to obtain these
interference fringes is the coherence of the light
source. The development of the laser in the 1960s
was a milestone in the production of highly coherent light sources and led to rapid progress in holography the ability to record and reproduce 3D images. Indeed holographic techniques for nano-scale
manipulation continue to be developed in this Department. In the second part of this experiment
you will learn how to produce your own holograms.
(b) its radius of curvature

(c) its refractive index
(d) the refractive index of water.
Figures F.2 and F.1 show the essential optical
arrangement to be used. The extended light source
is obtained from a sodium vapour lamp (a good
source of monochromatic light). The glass sheet
at 45 degrees reflects the light downwards onto a
plano-convex lens placed on an optically flat glass
surface. Light reflected upwards from the glass
surface (after it has passed through the glass lens
twice) can then be viewed with a travelling microscope placed immediately above the first glass
sheet.
to travelling
microscope
Extended
light source
glass sheet
at 45 degrees
Basic theory
In one dimension the amplitude of a travelling wave
at a point x, t in space may be expressed by:
glass surface
a(x, t) = a0 sin(t kx + )
spherical lens
(F.1)
where a0 is its maximum amplitude, is its angular

frequency, k its wavevector and its phase. It can
also be shown that = 2v/ and k = 2/.
Interference results when two or more waves arrive at the same point in space and time (the theory
can readily be extended to three dimensions) and
can be expressed simply in terms of the superposition (mathematical addition) of the amplitudes
a1 (x, t) and a2 (x, t) of the two waves. Hence if the
waves are exactly in phase their amplitudes add
constructively (to give a1 + a2 ), whereas if they are
exactly out of phase they will destructively interfere (to give a1 a2 ). Experimentally, we generally
record the intensity I(x, t) = a(x, t)2 ; it is this that
is stored on a photographic film or a CCD device.
Therefore in order to produce a stable interference pattern we need our interfering waves to have
constant amplitudes and phases. To achieve a constant phase relationship, our waves must be highly
monochromatic (single wavelength/frequency).
F1: Newtons Rings

The interference of light is used in a wide range of
applications, for example, in checking the quality of
lenses and mirrors used in cameras and telescopes.
In this experiment you will use interferometry to
determine:
Figure F.2: The optical arrangement for the Newtons rings experiment.
Figure F.3 shows (a much enlarged) diagram of
how the interference fringes are produced. Light
incident on the lower surface of the lens is partially
reflected (beam A) and partially transmitted. The
transmitted part gets partially reflected at the plate
glass and passes back through the lens (beam B).
These two beams are phase coherent with an optical path difference due to the air gap traversed
only by beam B. If the width of this gap at P is
dn then interference can occur producing bright or
dark fringes depending upon whether:
2dn = (n + 12 )0 ;
2dn = n0 ;
bright fringes
dark fringes
where n is an integer and 0 is the wavelength of

the light in free space. The extra +1/2 arises because of a phase change of for beam B on reflection from the plate glass, which is optically dense
compared to air (you should find out why in later
courses). Hence the fringes consist of a set of alternate dark and bright rings which are circular if the
lens surface is spherical and the plate glass is optically flat. The central fringe (zero path difference)
is dark and each fringe gives a contour of constant
air space thickness.
54
EXPERIMENT F.
OPTICAL INTERFERENCE
vernier scale. How easy is it to locate the centre of

the fringe? What errors are there in the measureGlass
ment of the fringe diameters? Determine the radius
Air or water
lens
of curvature of the lens and a rough estimate of
O
P
Glass
the error by plotting the appropriate graph in your
plate
laboratory book. Use the graph to see if it is necessary to take any more measurements to improve
Figure F.3: Ray diagram for the Newtons Rings the accuracy/precision of your results. Repeat the
experiment
process so that you have measurements for all three
lenses that were supplied with the apparatus.
A
If instead of air a liquid, whose refractive index

is , is used, then it has effect of reducing the wavelength of the light to = 0 / (you will see why in
later courses) so that the fringe conditions become:
2dn = (n + 12 )0 ;
2dn = n0 ;
bright fringes
dark fringes
It is a matter of elementary geometry that if the

lens has a long focal length and hence a large radius
of curvature r0 , then the radius of the ring, tn , (and
hence the distance OP) is given by:
t2n = dn (2r0 dn )
In practice dn 0.1 mm and r0 100 cm so dn
r0 and we get:
2dn = t2n /r0
Find the focal length of each lens. From this determine the refractive index of the glass lens and
a rough estimate of the error. Once you have obtained the radius of curvature of the all the lenses
you should move onto the measurement of the refractive index of water. Choose a lens to use based
on the difficulty and precision found in the previous
measurements.
Place a small drop of water on the plate and carefully place your chosen lens on it so that the liquid is uniformly sandwiched between the lens and
plate. Check the fringe pattern is circular, and use
the same procedure as before to determine the refractive index of the water with an estimate of its
experimental error. You may assume the refractive
index of air is 1.0. If you have time, repeat the
measurement for the other lenses so that you have
three different measurements of the refractive index
and their estimated errors. Are they consistent?
Hence the fringe conditions become:

t2n = (n + 12 )0 r0 ;
t2n = n0 r0 ;
bright fringes
dark fringes
The ring fringes with monochromatic light ( constant) are called Newtons Rings.
Experimental
The light source in this experiment is a sodium
vapour lamp. A lens is placed on the plate glass beneath the objective of the travelling microscope and
the light is directed onto the lens from above using
the strip of plane window glass as a partially reflecting mirror (see Fig. F.2). The interference fringes
are localized in space so the microscope must be focused on them. It is a good starting point to focus
a piece of paper on the glass stage, then to move
the whole lens beneath the objective lens until the
rings have been located. Look at the interference
fringes carefully. Are they circular? Is the central
spot dark? If not think of the possible reasons why
and make a note in your lab. book.
Once you have observed good fringes you should
determine their diameters 2tn using the coarse and
fine traverse screws of the travelling microscope.
Torches can be supplied to assist with reading the
Points to note
1. All optical surfaces must be clean and free from
dust. Check the rings are smooth and uniform
in shape and that the central one is dark.
2. Set the microscopes cross wire on the fringe
centre as precisely as possible and do not jog
the apparatus while making measurements.
3. This experiment works at the limit of precision
of the vernier scale on the microscope. Careful
reading of the verniers is essential.
Additional observations
While the apparatus is set up remove the lens and
liquid and clean the surface of the glass plate. Carefully place a clean microscope slide on the plate
and look at its surface. You should be able to see
interference fringes without needing a microscope.
What can you say about the flatness of the microscope slide on the plate? Take a soft pointer (e.g.
a pencil) and gently press down the surface of the
slide. What do you see? Can you explain it?
55
F2: Holography
If we illuminate a 3D object with highly monochromatic light and record an image on film we might
expect that we could reverse the process by illuminating the film with light of the same wavelength
to produce a 3D image of the original object. However, as noted in the first section, our image records
the intensity of the light rather than its amplitude.
In doing so it means that all of the phase information in the interfering waves has been lost and it is
this that is required to produce a correct 3D reconstruction of the original object. This is known as
the phase problem and has important consequences
in, for example, crystal structure determination by
x-ray or neutron diffraction (another way of describing interference).
The Hologram Experiment

Safety Notice:
Laser safety
The holography experiment uses a 1 mW
633 nm Class 2 laser. Staring directly into
the beam may cause permanent eye damage.
Observe the beam indirectly on a piece of
card. You should wait until your demonstrator has explained the safe use of the laser before you proceed with the experiment.
Chemical safety
Chemical solutions are used to develop and
fix the photographic film. Take care to avoid
splashing chemicals on skin. When using
these solutions you must wear gloves and a
lab coat. In case of contact with eyes, rinse
immediately with cold water. Please wash
your hands after this experiment.
It is possible to overcome this phase problem if

we record the light scattered from our object as it
interferes with that of a highly coherent reference
source. If we then illuminate this recorded image
with a highly coherent source (usually but not necessarily the same as the reference source) we ob- In this part of the experiment, you will create your
serve what appears to be a proper 3D image of our own holograms, using both the transmission and
original object. Although the basic principles of reflection methods.
holography were laid out in the 1940s it wasnt until the advent of the laser (which produces highly
coherent light) that holograms could be produced
Warning:
in a straightforward manner.
At this level a full explanation of how the holographic technique works is not possible, but we
can let you see how it works in practice. If you
wish to investigate further the principles should be
described in any good undergraduate textbook on
Modern Optics. You will produce holograms by two
methods; transmission and reflection. You will note
how both use the light from a single laser to illuminate the sample and produce the reference beam.
In order to produce good interference patterns
our photographic film must be capable of resolving
detail at scales on the order of the wavelength of
light (approx. 2000 lines/mm). Conventional film
used in pre-digital cameras typically has a resolution of over ten times lower (
100 lines/mm). Hence
we must use special film produced for the purpose.
Unfortunately this means the film also has low light
sensitivity (it takes longer to record the image) so
that we need long exposure or a high power laser.
For safety reasons we use a low power laser with an
output power of less than 1 mW. Experience has
shown that exposure times of the order of 5 minutes
are necessary to obtain suitable transmission images and 30-40 seconds for reflection images. Bear
in mind that everything must be stable to within
< /10 if the phase information is to be recorded
satisfactorily, so it is essential that everyone in the
room remains quiet and motionless while the film is
being exposed. Even the smallest nudge of a bench
can ruin an exposure.
The apparatus contains a front-silvered mirror and care must be taken not damage the
reflective coating by touching it or allowing
anything to touch the top surface.
Any optical components which are moved
must be secured thoroughly once they are in
their new positions. The mirrors in particular are top-heavy and are easily damaged.
The Transmission Hologram

This method, shown in Fig. F.4, is based on that
of Leith and Upatnieks. The helium-neon (HeNe)
laser beam is expanded by a diverging lens. The
beam splitter cube is a very efficient device using
multi-layer reflections to provide a near 50:50 division of the beam power. The reference beam intensity is then reduced (to lower the background
signal recorded on the film) by reflection from an
un-silvered mirror this is actually a flat plate of
glass, with the back painted matt-black to avoid
double-reflection. The other half of the beam is
reflected onto the object using a front-silvered mirror.
You may supply an object of your own choosing
for this experiment. Metal pencil sharpeners can
be especially successful.
56
EXPERIMENT F.
Beam
Splitter
Un-silvered
mirror (glass
plate)
Laser
Diverging
Lens
Reference
Beam
Film
Object
Scattered
Beam
Front-silvered
mirror
Figure F.4: The suggested experimental layout for transmission holography. (Above) Schematic view,
and (Below) Photograph of setup.
57
Experimental Procedure
Set up the apparatus as in Figure F.4, recording
the position of each element in your lab book:
1. Place the diverging lens and beam splitter
close to the exit aperture of the laser.
2. Position the mirrors in each expanded beam.
3. Place a piece of tracing paper in the film holder
and position the un-silvered, reference beam
mirror to fully illuminate the paper.
4. With the reference beam covered, position
your test object such that it scatters light onto
the tracing paper. Scattered light should be of
comparable intensity to the reference light.
5. Take care to ensure that the subject beam
doesnt fall on the reference mirror, nor the
film itself. It must only fall on the object.
6. The beam splitter, partial mirror, total mirror
and film holder should form the vertices of a
square approximately 20-25 cm to a side.
Your apparatus should now be aligned and ready
to produce a hologram. The following stages must
be carried out in either complete darkness or using
the green safelight.
13. Drop your film in the first tray on the left, the
developer. Push it under with print tongs and
agitate the solution by gently rocking the tray.
After 2 minutes, remove the film from the tray.
Holding the film up to a green safelight will
reveal a dark patch on the film. This indicates
a successful exposure as the crystals (grains)
of silver bromide in the emulsion are reduced
to black metallic silver.
14. Wash the film in the first waterbath for about
1 minute to remove all traces of the developer.
15. Drop your film into the third tray, the bleach.
Push it under with print tongs and agitate the
solution by gently rocking the tray. After 2
minutes, remove the film. 30 seconds into this
process, you can turn on the room light. However, this can only be done when the last
piece of film is safely in the bleach. The
bleach oxidises the silver image to silver bromide, the emulsion now becoming transparent
again with a diffuse appearance.
16. Wash the film for 1 minute in the final waterbath to remove the bleach. Change the water
if it appears contaminated.
17. Dry the film gently by rolling it in absorbent
paper. This removes water from the developed
and bleached emulsion structure. Complete
the drying by placing the film on a tray in the
print dryer for 1 - 2 minutes. Please turn
the print dryer off when not in use.
7. Ensure that everything you need is to hand

and that you are clear about what to do next.
Take care throughout not to move any optical
You can view the hologram using the arrangement
components. Turn the room lights off.
shown in Figure F.5. You may need to study the
8. Block the laser beam by placing a piece of film in different orientations to find the hologram.
polystyrene in front of it (the shutter). Do Remember that your choice of object may not have
not switch the laser off ; the output is more created a clear hologram, and excess vibrations
during the exposure may have rendered the holostable when left on.
gram useless. Good success can be achieved with
9. Remove a piece of film from the light proof bag the correct co-ordinated procedure. This works
in the film box. Close the bag and return it to best when everyone exposes their film at the same
the box. Place your film in the film holder, time and one individual is responsible for timing.
avoiding exposure to any stray light.
Developing should follow a production line process,
with one group developing as soon as the previous
10. Partially lift the shutter without exposing group has moved onto the next chemical bath. If
the beam. Hold it in the air for 20 seconds time allows, repeat the process with another object.
whilst table vibrations decay.
11. Lift the shutter out of the beam and expose The Reflection Hologram
the film for about 5 minutes. If necessary, ask
for a stopwatch. Do not lean on the table. Another method of producing holograms devised
in 1962 by Y.N. Denisyuk is both less sensitive to
12. After 5 minutes, replace the shutter and re- vibration than the two-beam method of Leith and
move the film from the holder.
Upatnieks and also allows the resulting holograms
to be viewed in white light. Holograms produced
With the room lights off, develop the film in by the Denisyuk method use photographic film with
the solutions provided as follows:
thick emulsions. The reference wave and the wave
58
EXPERIMENT F.
Hologram
Laser
Diverging
Lens
Figure F.5: Optical arrangement for viewing a transmission hologram.
reflected from the object reach the film from oppo- The following stages must be carried out in comsite directions, as shown in Fig. F.6.
plete darkness or using the green safelight.
The resulting interference fringes now form
3. Remove a piece of film from the light proof bag
sheets nearly parallel to the film surface with
in the film box. Close the bag and return it to
a separation of /2, giving about 50 planes in an
the box. Place your piece of film on the glass
emulsion of 20 m thickness. These planes act as
stage, avoiding exposing it to any stray light.
a diffraction grating, which, at a given viewing angle, effectively select only one wavelength from the
4. Flatten the film with another piece of glass.
white light for reflection.
The experimental arrangement is shown in Fig.
5. Place your object on the top piece of glass
F.7. You may need to adjust the height of the mirabove the mirror. Coins and keys work well.
ror box. As before, the helium-neon laser beam is
6. Stand back from the table and raise the shutter
expanded by the lens. The expanded beam strikes a
slightly. Allow 20 seconds for table vibrations
mirror at 45 degrees to a glass stage, passes through
to decay, then lift the shutter out of the path
the film onto the subject. Scattered light reflects
of the laser.
back onto the film where it interferes with the incoming beam.
7. Expose the subject for roughly 30-40 seconds
then replace the shutter.
Object
8. Remove and develop the film using the same
procedure given above.
Object wavefront
Holographic
film
Reference plane wavefront
View the hologram in reflection using a spotlight

for illumination. You will have to try a variety of
orientations to see your hologram. Some subjects
work better than others. Due to the low power of
the laser, the process has a shallow depth of field.
Flat objects such as keys and coins will work best,
because they are close to the film. Your hologram
will not seem very good until it is completely dry.
Some Properties of Holograms

The white-light viewable hologram you have made
Figure F.6: The Denisyuk direct beam method of acts like a filter to reflect only light of a wavelength
reflection holography.
corresponding to the spacing of the layer fringes
through the film. These fringes corresponded to
The experimental procedure is straightforward:
the red wavelength of the HeNe laser. However,
the bleaching process used caused the film to shrink
1. Check the optical alignment. With tracing paand shifted the spacing to shorter wavelengths.
per instead of film, ensure the expanded beam
The bleaching process converts the opaque silwill illuminate the film. The laser-mirror disver
grains into transparent silver salts, which have
tance should be approximately 50 cm.
a different refractive index to the surrounding gel
2. Block the laser beam with the shutter.
medium. This converts the absorption hologram to
59
Figure F.7: Direct optical arrangement for making a reflection hologram.
a phase hologram. The advantage of phase holograms is that they have a higher brightness and
contrast than absorption holograms.
Looking at the hologram in reflection is like looking through a window. Holograms exhibit parallax.
As you move your head from side to side, different
parts of objects towards the back appear or disappear behind objects in the foreground.
this distance compare with the object to film

distance in the set-up to record the hologram?
Transmission holograms made by Leith and
Upatnieks method have the property of redundancy. Optical information about each point
on the object is spread out over the whole of
the hologram. This means that any piece of
the hologram has information about the whole
object. You can confirm this by cutting one of
your holograms in half. What do you observe?
In contrast, reflection holograms made by the
Denisyuk method with the object close to the
emulsion have optical information about any
one point of the object only in a small region
on the emulsion near to the object point.
By considering the convex nature of the layer

fringes when viewed in reflection, it can be
seen that a virtual image is produced behind
the hologram. Try viewing the hologram in
transmission by looking through it towards the
spotlight. What sort of image do you see?
Viewing the hologram in transmission so that
the object appears behind the film, use the
It is interesting to note that if a lens is present
parallax method to determine the distance of
as
part of the object, it still acts as a lens of the
this virtual image behind the film.
same focal length in the holographic image. Such
There is also a real image associated with holographic lenses are used in the bar-code readers
the hologram which can be viewed by plac- at the check-out counter of supermarkets. Holoing a screen at its location. With light passing grams are also used to record small displacements
through the hologram so that the virtual im- of a surface or small changes in refractive index.
age can be seen in transmission, try finding
the real image by placing a film holder carrying a piece of tracing paper in contact with
Formal Report Guidance:
the film on the side away from the light source.
The Holography experiment must not be writSlowly move the tracing paper away from the
ten up as a formal report, but you are may
film until you see a sharp image. Measure the
write about the Newtons Rings experiment.
distance from the film to the paper. How does
Experiment K
The Black Body Radiation Law

Aims:
1. To plot black-body radiation intensity as a function of wavelength for a tungsten filament
lamp at different temperatures;
2. To find the ratio of h/k, where h is Plancks constant and k is Boltzmanns constant;
3. To demonstrate the Wien displacement Law.
Objectives:
1. To learn to set up and calibrate an optical spectrometer;
2. To understand the operation of the lock-in amplifier.
chopper
condensing
lens
prism
telescope
collimator
lamp
pyroelectric
detector
chopper
controller
spectrometer
lock-in
amplifier
Figure K.1: The spectrometer for studying black-body radiation.
Further reading:
Company, New York, 2004). Chapter 20-4.
60
61
Introduction & Overview
Relative Intensity
3000 K
In this experiment you will first set up and calibrate a spectrometer system by using a known light
source with well-defined emission lines. Once the
system is calibrated you will investigate the emission spectrum of a tungsten bulb which approximates a black body.
This experiment requires patience and precision
to set up properly. In the first week you should
spend some time learning to set up the apparatus carefully and making notes on the procedure.
Once you have obtained an emission spectrum you
should insert the tungsten bulb and obtain the data
required for Part A of the analysis. This should
be analysed before you attend labs for the second
week.
In the second week you will need to quickly recalibrate your system using the procedure you noted
in your lab book, then obtain black body spectra
from the tungsten lamp at a number of temperatures.
2500 K
Theory
The electromagnetic radiation in any enclosure in
thermal equilibrium is isotropic, unpolarised, and
has an energy density and frequency distribution
that depends only on the equilibrium temperature
T not, for example, on the material forming the
walls. If this were not the case, the equilibrium
temperature of an absorbing body placed in the
enclosure would not in general be equal to T .
The radiation may be shown to be the same as
would be emitted by any black body at the equilibrium temperature T . Since its properties are independent of the precise nature of the emitter, blackbody radiation is of fundamental theoretical interest. Historically, it is important because it defied
classical explanation, and its study ultimately led
Planck, in 1901, to propose that electromagnetic
energy existed only in discrete quanta, each with
energy h, where is the frequency.
Plancks hypothesis led to an expression for the
energy density, (), of black-body radiation at
temperature T :
2000 K
1000 K
1500 K
1000
2000
3000
4000
5000
Wavelength (nm)
Figure K.2: Black-body spectra at different temperatures, as given by Eq. K.1. The maxima, as
given by the Wien displacement law (Eq. K.2) are
indicated.
Hence a black body at say 1000 K will emit its maximum power per unit wavelength in the infrared
at a wavelength of 2.9 m. The appearance of
the black-body spectrum at various temperatures
is shown in Fig. K.2. The peak lies in the infrared
region of the spectrum for all temperatures accessible with a tungsten filament lamp. On the other
hand, solar radiation peaks in the middle of the
8hc

.d
(K.1) visible spectrum, corresponding to a temperature
().d = 5
hc
exp kT
1
of around 6000 K.
At any wavelength max it can be seen that:
The factor d appears on both sides of Eq. K.1,
because any measurement of () will be multiplied
exp(hc/kT ) 1
by the aperture of the instrument (see below).
The wavelength max at which () has its maxi- This means that, on the short-wavelength side of
mum value is found to obey the Wien displacement the maximum, it is a good approximation to write:
law:
() 5 exp(hc/kT )
(K.3)
max T = 2.898 103 m.K
(K.2)
62
EXPERIMENT K. THE BLACK BODY RADIATION LAW
oscilloscope
out
condensing
lens
collimator
chopper
controller
in
lockin out
amplifier
ref.
Ch.1
Ch.2
beam
chopper
to lockin
amplifier
entrance slit
lamp
(mercury or tungsten)
goniometer
table with
prism
telescope
detector
or eyepiece
Figure K.3: The arrangement of the black-body radiation experiment.
The Experiment
wavelengths are summarised in Fig. K.4. Once the

relationship between spectrometer angle and waveMeasurements are to be made of the wavelength length is known, the mercury lamp is replaced with
distribution of the radiation emitted by a tungsten a tungsten lamp, whose output approximates the
filament lamp. The apparatus is shown in Figures black-body radiation shown in Fig. K.2. The major part of the experiment involves measurements
K.1 and K.3.
of this spectrum at different temperatures.
Radiation from the lamp is focused on the entrance slit of the prism spectrometer. The beam
is dispersed by the prism and then focused by the
Experimental Setup
telescope objective onto a second slit, beyond which
is an infrared-sensitive pyroelectric detector (lead
zirconate-titanate), in which the slight temperature
Safety Notice:
change due to the absorbed energy gives rise to electrical charges on the crystal surface. Infrared or
Do not try to view the mercury spectrum
visible radiation passing through the chopper and
with the naked eye - it will be uncomfortably
incident on the detector gives rise to an output curbright. Instead, project the spectrum onto a
rent proportional to the intensity. The chopper is
piece of paper.
necessary, because the detector is only sensitive to
It is highly unlikely that the mercury lamp will
a changing signal. The useful wavelength range of
break. However if it does you must leave the
the detector is limited by the glass optics of the
area immediately, evacuate the room, close
spectrometer, which absorb radiation outside the
the doors and tell the laboratory technician.
approximate wavelength range 0.35 2.2 m.
After passing through appropriate electronic amplification, the signal is fed to a meter. A lockin amplifier is used for this; the lock-in amplifier
only amplifies the signal modulated by the chopper,
rejecting any ambient background signal. When
correctly adjusted, the lock-in amplifier produces
a voltage output proportional to the intensity of
radiation falling on the detector. For convenience,
and to aid the initial set-up process, this signal will
be viewed and measured using an oscilloscope.
The experiment will be divided into two main
parts. First of all, it is necessary to set up and
calibrate the spectrometer. This is achieved using
the characteristic lines from a mercury lamp, whose
Setting up the spectrometer

Figure K.3 shows a suitable arrangement of the
spectrometer. Note: for satisfactory measurements
with the infrared detector and either source it is essential that as much light as possible passes through
the system. Set up the apparatus as follows:
(a) Place the mercury lamp in line with the collimator, and adjust their relative positions to
ensure the light is passing straight through the
collimator - you will see a vertical line centred
on the exit lens of the collimator when they
are correctly positioned.
63
(b) Insert the condensing lens: position it so that
an image of the source is focused on the entrance slit. Hold some paper in front of the
prism and slowly adjust the lens to achieve a
symmetrical circular image on the paper. The
exit lens for the collimator should not be extended.
(c) The collimator slit width may need adjusting;
if the slit is too wide, the resolution of the
optics will be poor. If the slit is too narrow,
the intensity will be too low.
(d) The collimator is now delivering parallel light
to the prism. The prism should be mounted
centrally on the goniometer table and locked
in place. Rotate the goniometer table until
the yellow mercury line is visible (projected
through the telescope onto a piece of paper),
and then carefully adjust it so the angle of
deviation, , for the yellow line is minimised,
whilst still remaining focused. Next, lock the
goniometer table in position for the remainder
of the experiment.
(e) Observe the image on a piece of paper and ensure the spectral lines are centred horizontally
within the telescope. If they are not, use the
three adjustment knobs on the prism table to
level the prism.
Setting up the detection system
Important Notice:
The detector is delicate and can be easily damaged. Do not twist the wire either at the detector end or the metal case - the internal
wires will disconnect and the repair is timeconsuming. Insert or remove the detector in
the telescope smoothly without rotating it.
The detection system consists of a pyroelectric

detector, a beam chopper with drive electronics, a
lock-in amplifier and an oscilloscope. Follow these
steps to ensure the detector is operating correctly.
(a) Remove the eyepiece, and replace it with the
detector, making sure that the yellow line falls
exactly on the vertical detector slit.
(b) Set the chopper frequency to 15 3 Hz.
(c) Connect the detectors voltage supply (9 V
battery), and connect the detector output to
the signal input on the lock-in amplifier.
(d) Connect the chopper driver output to the reference input on the lock-in amplifier, and to
Ch. 1 of the oscilloscope.
(e) Connect the lock-in amplifier output to Ch. 2
of the oscilloscope.
(f) The operation of the lock-in amplifier is quite
complex, and is summarised in Appendix A.
Initially, set the chopper driver controls to:
(i) EXT/INT set to INT
(ii) 10S/1S set to 1S
and the lock-in amplifier controls to:
(i) Sensitivity set to 300 V initially
(ii) Time Constant set to 1 ms initially
(iii) X Offset 0 V
(iv) Y Offset 0 V
(v) Offset OFF
(vi) Output Select R
(vii) Phase Shift 0 on both switches
(viii) 1F/2F switch 1F
(g) Observe the signals on the oscilloscope. The
chopper reference signal should appear as a
square wave, and the amplifier output, R, as
a periodic signal. Use the measure function
on the oscilloscope to display the mean voltage
value of the signal on Ch. 2. This is essential
as you will depend on this reading when you
take your measurements.
If you cannot see a signal on Ch. 2, try adjusting the telescopes position slightly. If this
fails, try altering the input sensitivity of the
amplifier or the oscilloscopes voltage range.
If you can see the mean voltage on the oscilloscope change when the beam is blocked your
detector should be functioning correctly - the
system probably needs to be re-aligned. If the
mean voltage does not change, check the voltage of your battery.
(h) Once there is a clear signal on Ch. 2, increase
the time constant on the lock-in amplifier until the periodic variation is smoothed out to a
flat trace, offset from the origin. A time constant of 1 s should suffice, though for weaker
spectral lines 300ms will offer greater visibility of changes. The mean voltage displayed
on the oscilloscope is now proportional to the
input intensity.
N.B. Occasionally the Ch. 2 signal will show
a marked increase when someone touches the
spectrometer. If this happens, ask a staff member how to ground the kit.
64
(i) The detection system should now be opera- should appear similar to the one in Fig. K.5. Note
tional. Verify this by slowly scanning the tele- that the absolute values of the angle are not important as the initial orientation of the goniometer
scope on and off the spectral line.
table is arbitrary. Although this plot is initially inEach detector was mounted inside its metal can by
tended as a wavelength calibration, later you will
hand and each slit was handmade by spacing two
use its gradient to correct the detected intensity.
semi-circular pieces of brass with metal foil then
soldering them in place. Therefore there will be
subtle variations between detectors on account of Notes on detecting spectral lines
their distance from and angular orientation wrt the
When using the mercury vapour lamp, it should
slit.
be possible to optimise the lock-in sensitivity and
oscilloscope voltage range to yield a reading of a few
Spectrometer Calibration
volts when detecting the green or yellow spectral
lines. The background reading (when the mercury
Before you measure the spectrum of the tungsten lamp is blocked) should be several tens of millivolts.
lamp, the spectrometer must be calibrated. This is
achieved using the characteristic emission lines of (a) Traverse the telescope very slowly (the ammercury, the most prominent of which are given in
plifier has a 1 second response time, so it is
Fig. K.4 (take care not to confuse them). With the
easy to miss a spectral line). For weaker lines
detector in place, the intensity distribution over the
and the IR series, it may be beneficial to reduce
visible and infrared parts of the spectrum may be
the lock-in amplifier time constant to 300ms.
examined by rotating the telescope.
You will be less likely to miss a line, though
Take some preliminary measurements in the visithe detector voltage will not be flat.
ble region prior to the main calibration, projecting
the spectrum onto paper and noting the angular (b) It is not usually possible to detect all 5 infrared lines (the 1530 nm line is very weak).
positions of the various visible lines.
Therefore, plot a graph of voltage versus anThe fixed detector slit gives adequate resolution
gle as you go, so that you can work out which
in the infrared. Furthermore, it should be possible
wavelengths you have detected.
to obtain an adequate signal in the visible spectrum
with an entrance-slit width where green and yellow
(c) The infra-red lines are close together - you will
lines are resolved easily. In the calibration it is
not get a zero intensity reading between them.
important to identify the lines with wavelengths
greater than 1000nm. Record the intensity of each
line and its angular position, and use it to help Fitting the calibration data
identify the wavelength from Fig. K.4.
Ultraviolet light is filtered by the glass of the One of the biggest difficulties in performing this
bulb and will not be present in the spectrum. You experiment is in accurately measuring the light inare unlikely be able to find blue-green, red or the tensity, and hence determining the positions of the
1530nm IR lines, as they are too weak. When look- lines in the mercury spectrum.
Write down estimates of the errors in the posiing for these lines, try decreasing the lock-in time
tions of the mercury lines, explaining how you arconstant.
rived at your figures.
The calibration of your spectrometer is crucial in

accurately
determining the wavelength correspond 1
ing to a given spectrometer angle, and later you will
need to use it to determine the intensity correction
factor at each spectrometer angle.
To accurately convert your angles into wave 2
lengths, you will now fit a suitable curve to the
experimental points, then use this curve to interpolate and extrapolate to other wavelengths and
angles. The optical dispersion of glass is often de 1
scribed using the Cauchy expression:

2
Figure K.5: An example of a calibration curve.
n() = A +
C
B
+ 4
2
(K.4)
Plot a calibration curve of spectrometer angle where A, B and C are constants for the material.
versus wavelength for the mercury lines. The curve For small deviations, the angle of deviation goes
65
red
Intensity (not to scale)
Ultra-violet
deep violet
blue
blue-green
green
yellow
Infrared
365.0
404.7
435.8
491.6
546.1
578.0
690.5
709
1014
1129
1360
1530
1707
Wavelength (nm)
Figure K.4: The line-spectrum from the mercury lamp.

approximately as n, so it is reasonable to write:
() = A0 +
C0
B0
+ 4
2
(K.5)
where A0 , B 0 and C 0 are constants to be found.

Using Origin, fit your calibration data to a curve
of the form given by Eq. K.5. You will need to define the fitting function. There is an Origin tutorial
on the DLM, or staff members can help.
What are the values and the errors obtained from
your fit for A0 , B 0 and C 0 ? Does the fitted curve
pass through all of the error bars?
Make sure you take the required data for
part 1 of the tungsten data before you attempt the remaining questions.
From your calibration curve:
(1) Deduce an equation for the calculation of the
wavelength.
(2) Is the error in the wavelength a constant value,
or does it vary with spectrometer angle?
Replace the mercury lamp with the tungsten lamp,

which emits a spectrum approximating that of a
black body. A final alignment process is necessary
before you can proceed. Do not rotate the prism or
goniometer table! Instead, adjust the positions of
the lamp and condenser to focus the light onto the
entrance slit. Take care to place the lamp at the
same height as the slit since the lamp assembly will
quickly be too hot to touch. Rotate the telescope
to the yellow part of the spectrum and set the lamp
power supply to 12 V ( 3.9 A).
The tungsten lamp only illuminates a small part
of the entrance slit. Ensure your signal leaves the
telescope centred horizontally. If it does not, try
adjusting the height of the condensing lens (and
thus the image of the filament) to maximise the
detector output.
Insert the detector and scan the telescope across
the infrared region. If you observe a peak signal of
between 2 V and 14 V (depending on the particular
detector) you are ready to record spectra. If the
lamp is knocked at any time, the signal may
be lost.
(3) Are there systematic errors involved?

(4) Discuss the roles of systematic and random errors in the determination of wavelength.
Experimental Setup 2 - Aligning the tungsten lamp

Safety Notice:
When the tungsten filament lamp has been
switched on for any length of time the whole
assembly (stand, clamp, boss and shroud) becomes extremely hot. Take care not to burn
yourself. At the end of the experiment, be
sure to switch this lamp off.
Black-Body Data Collection

The aim of the experiment is to measure the intensity of the radiation from the tungsten filament
as a function of both wavelength and temperature,
thence to compare the results with the Planck formula, given in Eq. K.1.
The temperature of the tungsten filament may
be found from its resistance, R, using the empirical
relation:
R = R0 (1 + 5.2 103 T + 4.5 107 T 2 )
where T is the temperature in C, and R0 is the
resistance at 0 C. R may be found from measurements of the current through and voltage across
the lamp. The value of the filament resistance at
room temperature, R20 , is 0.2 .
66
For wavelengths beyond about 2 m there is sig- corrected using the calibration curve, to obtain the
nificant attenuation in the glass optics and the de- Gc () which is appropriate to a fixed , and is
tected intensities will not be representative of a thus proportional to () in Eq. K.1.
black body. However, for filament temperatures
For small and , the corrected value is:
above 1500 K you should be able to observe the
peak in the Planck distribution and its shift to
Gc () = G().(/)
shorter wavelengths with increasing temperature,
as depicted in Fig. K.2.
where / is the gradient of your calibration
curve. Origin can differentiate the curve - ask your
Tungsten Data Week 1 - Intensity ver- demonstrator for advice.
sus Temperature
How accurately can you determine the intensity

correction factor from the gradient of your graph?
For a given wavelength, max , the detected

In principle, further corrections are required for
intensity G() may be assumed to be proportional
to () and should therefore vary with temperature the variation with wavelength of:
as:
G(, T ) exp(hc/kT )
(a) the filament emissivity;
A plot of ln(G(, T )) versus 1/T should therefore

produce a straight line, from whose gradient the (b) the detector sensitivity;
fundamental constant h/k can be calculated.
Take the appropriate data and complete this (c) the effective spectrometer aperture.
analysis before returning to the lab for the second
It is reasonable to assume that the filament emisweek of the experiment.
sivity is roughly constant, and the pyroelectric deTungsten Data Week 2 - Intensity ver- tector sensitivity is stated to be wavelength independent over the range covered in the experiment.
sus wavelength
If the spectrometer is set up correctly, so that all
Before starting the data collection with the tung- the radiation of the required wavelength that leaves
sten lamp in week 2, re-align and calibrate your the collimator reaches the telescope, then the aperspectrometer. (The fitting process can be completed ture should not change as the wavelength is altered
by rotating the telescope, so (c) should be constant.
after the lab.)
Hence these further corrections are not likely to be
Working at a series of fixed temperatures besignificant.
tween 1500 K and 3000 K, measure the spectral inThe dispersion correction described above will
tensity from the visible region through to the infrared (i.e. from about 400 nm to above 2 m). At also affect the position of the peak in Gc () for
each temperature, you should be able to observe a each filament temperature. Plot graphs of your corpeak in intensity as predicted by the Planck for- rected intensities, Gc (), versus wavelength. How
well do the corrected positions of the peaks obey
mula (Eq. K.1).
Make detailed measurements of the detected sig- Wiens law?
nal, which we will call G(), as a function of waveOnce the corrected voltages Gc () have been callength for at least four or five different fixed temper- culated, then for max we have from Eq. K.3:
atures. However, the relation between G()
and () is itself wavelength dependent, so a
Gc () 5 exp(hc/kT )
correction factor is needed before any comparisons may be drawn.
so that at a given filament temperature the variation with 1 of the quantity ln(5 Gc ) should be
linear, and yield an experimental value for the ratio
Intensity correction
h/k.
The difficulty here is that your measured curve
G() reflects not only () but also the changing
Plot your data for max for each temperdispersion of the prism spectrometer. As Fig. K.5
ature, as described above.
illustrates, the curvature of the dispersion curve
means that a given detector slit width, , corre What values do you obtain for the ratio h/k?
sponds to different values of , the range of detected wavelengths, in different parts of the spec Can you think of any systematic errors which
trum. This means that the observed voltage output
might affect this result?
G(), which corresponds to a fixed , needs to be
67
Results & Discussion
Appendix C
Quote your results for Plancks constant from

both methods, giving a careful explanation of your
quoted error in each case. (Hint: think carefully
about the relationship between the tungsten filament resistance and temperature).
Comment on the biggest sources of error in the
experiment, both random and systematic. How
might they be reduced?
Which is the more accurate method of doing this
part of the experiment?
The Lock-in Amplifier
1. Taking values of G(, T ) from the graphs of

intensity as a function of wavelength by interpolation;
2. Making direct measurements of G (T ) by
varying filament temperature at a fixed wavelength.
A lock-in amplifier is a clever device aimed at detecting low signal levels. It operates by multiplying
the input signal, Uin , by a reference signal e.g. a
sine wave at the frequency of interest. The output
signal, Uout , is obtained by integrating this product over a pre-determined time period, , which
should ideally cover many cycles of the reference
frequency. The orthogonality of sine functions ensures that the integral of a product of two sines will
be zero provided that:
1. the two sines are of different frequency;
2. the time constant, , is sufficiently long.
Thus, the following integral:
Z
1 t
sin(ref t0 + ).Uin (t0 ).dt0
Uout (t) =
t
Why? How does your new value for h/k compare

with that obtained at a constant wavelength? Are
produces an output, Uout , which is proportional to
there any different systematic errors?
the amplitude of the input signal chopped at ref .
Most importantly for our application, Uout will be
zero for signals (including DC signals) at all other
frequencies, which means that we can safely operate this equipment with the lights on in the lab.
This also means that most of the noise in the system is rejected, as the thermal and electrical noise
will possess a wide range of frequencies, most of
which will not be detected. This makes the lock-in
amplifier useful under poor signal-to-noise conditions.
The lock-in amplifier used in experiment K is
able to measure both the in-phase, X, and outof-phase, Y, components of the input signal, relative to the phase of the reference. The constant
in the above equation is an additional phase constant that may be introduced. In K we do not use
the phase information, and instead use the magnitude, R, which is determined from X and Y using
Pythagoras.
Ideally, the time constant, , should be set as
long as possible to ensure that the integral averages
to zero. In experiment K we use = 1 s, so that
we integrate over ref (i.e. around 15) cycles. To
improve the noise rejection, we should really use
a larger , but this would make the response of
the detector too slow. The alternative option, to
increase ref , reduces the sensitivity of the detector
and is not recommended.
Experiment M
Measurement of the Charge of the

Electron
Aims:
1. To use the effect of an electric field on small charged oil droplets to measure the charge of
an electron.
Objectives:
1. To gain experimental technique in these practically challenging measurements;
2. To develop experimental judgement in choosing the most suitable data collection methods
Figure M.1: The Millikan apparatus for measuring the charge of an electron, showing the computer,
video camera and mounted Millikan chamber
Further reading:
Halliday, Resnick, & Krane, Physics Volume 2, p 617.
Richtmyer and Kennard, Introduction to Modern Physics.
68
69
Safety Note
Radioactive sources: The ionization source
in this apparatus is Thorium-232 which is a
naturally occurring, low level alpha-particle
emitter with a half-life of 1.41 1010 years.
It is not regulated in its use and poses no
hazard to the user.
High voltages: This equipment uses high
voltages. Ensure the high voltage power supply is switched off whenever you connect up
the apparatus or modify it. The power supply
is protected so that it cannot deliver enough
current to be lethal, but an electric shock
might hurt you or damage the apparatus.
Make sure that you connect the high voltage
supply to the plate voltage connectors and not
to the thermistor connectors.
where r is the radius of the drop and is the coefficient of viscosity of the air. Given , , and g,
the value of ne can be extracted from the terminal
velocities in two different fields.
We may, for example, measure:
(a) the field E0 for which the drop remains stationary (v = 0), and;
(b) the terminal velocity v0 in the absence of a
field (E = 0).
For (a) we have F1 = 0, so that:
ne =
gf M
E0
and for (b), F1 = F2 , so that using M = 4r3 /3

to eliminate r from eq. M.2 we obtain:

3/2
v0
M = 9 2
(M.4)
gf
Hence the charge carried by the drop is:
s
9 2 3 v03
ne =
E0
f g
Introduction
In this experiment the electronic charge e is to be
measured by essentially the same method as that
first used by Millikan in 1909. Millikan spent most
of his career researching cosmic rays, a term which
he invented. He was a firm believer in the atomic
theory or granular representation of nature. Millikans oil-drop experiment was largely conceived to
prove his belief in the fixed value for the electronic
charge. The experiment is, as you will discover,
fraught with difficulties.
(M.3)
(M.5)
The magnitude of e emerges from measurements on

a large number of drops. Since there is no way of
telling whether a drop has gained or lost electrons
in acquiring its charge, the experiment does not tell
us whether e is positive or negative.
Apparatus
Small drops of oil are produced as a fine spray

ejected from an atomiser. During this process the
drops may be stripped of or accumulate electrons.
The resulting number of electrons on such a drop
is not predictable but usually differs from that required for electrical neutrality by a small integer n
of either sign.
If a drop carrying n extra electrons finds itself
in a region of electric field E, it experiences a force
neE. With E directed downwards, the total downward force on the drop at rest is then:
F1 = neE + M gf
(M.1)
where M gf = M g( )/ is the apparent weight

of the drop, of density , surrounded by air of density (Archimedes Principle). The drop accelerates until it reaches a terminal velocity, v, at which
F1 is balanced by the viscous force F2 given by
Stokes formula:
F2 = 6rv
Figure M.2: The key components of the Millikan

apparatus.
The apparatus is shown in Figs. M.1 and M.2. The

main difference from the apparatus used by Millikan is that there is the opportunity to use a webcam and computer to record and analyse the mo(M.2) tion of the oil droplets.
70
EXPERIMENT M.
MEASUREMENT OF THE CHARGE OF THE ELECTRON
Figure M.3: The important parts of the apparatus platform.
The primary aim of the experiment is to obtain a

value for the electronic charge by making the measurements by eye in the original manner. If you
complete this analysis within the lab time you may
obtain a webcam from the lab technicians and use
it to investigate and evaluate what improvements,
if any, can be made by using modern technology.
The main component parts of the Millikan apparatus platform are shown in Fig. M.3. In particular,
note the following:
ionization source lever (with Ionization ON,

Ionization OFF, and Spray Droplet positions);
plate charging switch (on a 1 meter cord to
prevent platform vibration during switching).
Details of the droplet viewing chamber are given

in Fig. M.4. Note the two brass capacitor plates,
which define the high-voltage chamber, and the
plastic spacer, 8 mm thick, which sets the distance between them.
Oil droplets are introduced into the space be droplet viewing chamber (see details below);
tween the capacitor plates, across which a poten viewing scope (30, bright-field, erect image) tial difference, variable from 0 to 500 V , may be
with reticle (line separation: 0.5 mm major applied from the power supply. The direction of
divisions, 0.1 mm minor divisions), reticle fo- the field between the discs may be reversed using
cusing ring, and droplet focusing ring;
the plate charging switch.
lamp housing for light pipe with an infra-red

heat absorbing window;
Initial Setup
thermistor connectors (thermistor is mounted

in the bottom plate);
Measure the plate separation
thermistor table (resistance vs. temperature);
plate voltage connectors;
focusing wire (for adjusting viewing scope);
Disassemble the droplet viewing chamber by lifting

the housing straight up and then removing the upper capacitor plate and spacer plate (see Fig. M.4).
Measure the thickness of the plastic spacer with a
71
Figure M.5: Insertion of the focusing wire into the

top capacitor plate.
viewing scope. Then return the focusing wire to its
storage location on the platform.
Measure the chambers temperature

Connect a multimeter to the thermistor connectors
and measure the resistance of the thermistor. Refer to the Thermistor Resistance Table (Table M.1)
to find the temperature of the lower brass plate.
The measured temperature should correspond to
the temperature within the droplet viewing chamFigure M.4: Exploded view of the components in
ber. Check the temperature periodically during the
the droplet viewing chamber.
experiment as it may rise due to heating from the
lamp.
micrometer. Use care when handling the brass
plates and plastic spacer to avoid scratching
them.
Focus the viewing scope

1. Reassemble the plastic spacer and the top capacitor plate onto the lower capacitor plate.
Replace the housing, aligning the holes in its
base with the housing pins (see Fig. M.4).
Note: The thorium source and the electrical
Table M.1: The thermistor resistance at various
connection on the lower capacitor plate fit into
temperatures.
appropriately sized holes on the plastic spacer.
2. Unscrew the focusing wire from its storage
place on the platform and carefully insert it
into the hole in the centre of the top capacitor
plate (see Fig. M.5).
3. Bring the reticle into focus by adjusting the
reticle focusing ring.
4. View the focusing wire through the viewing
scope, and bring the wire into sharp focus by
adjusting the droplet focusing ring.
The main experiment:

Introducing droplets into the chamber
1. Prepare the atomizer by rapidly squeezing the
bulb until oil sprays out. Ensure the atomizer
tip is at 90 to the shaft (Fig. M.6).
Focus the halogen light source
2. Move the ionization source lever to the Spray

Droplet Position to allow air to escape from the
chamber during the introduction of droplets
into the chamber.
Adjust the brightness of the lamp to a suitable

level, whilst viewing the focusing wire through the
3. Place the nozzle of the atomizer into the hole

on the lid of the droplet viewing chamber.
72
EXPERIMENT M.
MEASUREMENT OF THE CHARGE OF THE ELECTRON

5. Record the times and distances for each measurement. The observer should not look away
from the viewing chamber or they will risk losing the droplet.
6. From a plot of distance versus time, calculate
the terminal velocity of each particle (if terminal velocity has actually been attained).
Figure M.6: Correct position of the atomiser tip.
Notes
The success of the experiment depends consider4. While observing through the viewing scope,
ably on the skill of the experimenter. It will be an
quickly squeeze the atomizer bulb once. Then
advantage to pay attention to the following points:squeeze it slowly to force the droplets through
the hole in the droplet hole cover, through the (a) The oil drops will be hard to see if the illuminadroplet entry hole in the top capacitor plate,
tion is inadequate. Adjust the background iland into the space between the two plates.
lumination to get good contrast. Ask a demonstrator to help if you have undue difficulty.
5. When you see a shower of drops through the
viewing scope, move the ionization source lever (b) Large drops often carry large charges, which
to the OFF position.
cannot be measured accurately enough to establish e unambiguously. Although they are
more difficult to see the smallest drops are the
best ones to measure, as they usually carry
Recording and measuring droplets
only a few excess charges. It is the integer ratio of these charges that allows the value of a
When you can observe the droplets on the screen,
single electronic charge to be measured.
apply a voltage to the capacitor plates. Many drops
will keep falling, but some will slow or move up(c) Convection currents must be avoided. The mowards. Adjust the voltage such that you hold one or
tion of the drop is to be observed in still air.
more droplets completely stationary. If no droplets
respond in this way, move the ionization source (d) The velocity v0 is the terminal velocity of the
lever to the ON position for about 5 seconds,
drop. Are special precautions necessary to ento produce some charging.
sure that this velocity is reached before meaThe motion of the drops is to be recorded and
surements are taken?
measurements of their velocity may be made by
timing them as they fall through the field of view. (e) Make sure that you observe the same droplet
stationary and under zero field. A moments
Measurements on at least twenty drops are usually
inattention may cause much time to be wasted.
needed before a value can be assigned unambiguously to e. Look at your results as you go through
the experiment to see if you can identify the inteData Analysis
ger steps in n for the charges on the drops. The
density of the oil is 886 kg m3 ; for , and g use
This section describes a thorough data analysis proKaye and Labys tables.
cess, including detailed error calculations. It should
To collect the measurements, proceed as follows: be completed for data taken by eye and then camera if possible so that you can compare the advan1. Record the voltage for which your droplet re- tages and disadvantages of the two methods.
mains exactly stationary on the screen.
(1) What is the error in the voltage across the
2. Adjust the focus to ensure that your droplet
plates in the apparatus?
appears sharp this is important if you are to
trust the reticle calibration.
(2) What is the error in the distance between the
electrodes?
3. Switch off the high voltage supply.
(3) Hence determine an estimate for the fractional
4. Using the reticle on the telescope determine
error in electric field within the cell.
the terminal velocity of the drop as it drops.
How do you decide whether it has reached ter- (4) For each droplet, draw a graph of its position
minal velocity?
versus time during free fall. You should:
73
(a) include error bars for both vertical posi- telescope, place the camera in front of the teletion and time;
scope and start the Logitech Webcam software on
(b) find the linear part of the graph and use the computer so that you can see an image of the
its slope to determine the terminal veloc- droplets on the computer monitor. You will need to
disable the automatic focus once the initial image
ity;
is obtained, and rely on the manual focus onscreen
(c) estimate the error in the terminal veloc- and on the telescope for fine adjustments.
ity, graphically or otherwise.
Adjust the webcam settings until the best image
(5) Estimate your uncertainty in , and arising of the droplets is obtained - you may need to adjust
from your uncertainty over the exact temper- the zoom, brightness and contrast in the Advanced
Settings window.
ature and humidity within the Millikan cell.
You should now be able to record videos of the
droplet motion which you can then view in QuickFor each oil drop, the charge is given by:
time. When you take data you should ensure you
s
3
3
record the filename and the conditions under which
9 2 v0
(M.6) it was recorded (voltages etc...); you will need this
ne =
E0
f g
to analyse the data.
Once you have recorded a movie, play it back
(1) Find the fractional error in the drops charge.
Which variable makes the largest contribution frame by frame, using Quicktime. You can use the
Movie Inspector option to find time stamps for
to the error?
each frame, and you should measure the number
(2) Hopefully, the error in the charge of each of pixels the droplets fall (e.g. using PixelRuler)
droplet will be ne. Is this the case?
then convert this to a distance.
With this data you should be able to plot a dis(3) By looking for integer, or near integer, ratance
versus time graph. Can you tell from this
tios between the charges of pairs of droplets,
data
whether
terminal velocity has been achieved?
it should be possible to unambiguously assign
If
so,
calculate
the terminal velocity.
values of n to each droplet.
Once you have established the method, record
(a) Do this, and then plot a graph of droplet several drops and use the data to calculate the
charge against n, including error bars.
charge on the electron.
(b) Measure the slope of the graph and determine its error, to find your final estimate
for the charge on the electron, e.
The extended experiment:

The technicians have webcams available which you
may borrow after you have completed the full analysis for the data taken by eye. A simple extension
to the experiment is to use a camera to record the
motion of the droplets within the chamber and obtain some reliable data to determine the terminal
velocity of the droplets, and how long the droplets
take to reach this velocity.
If you have sufficient time, you may also determine a second value for the electronic charge, this
time using the camera to record the motion of the
droplets, with the aim of determining whether the
quality of the results may be improved by using
more sophisticated techniques.
Setup
Set up the chamber the same way as previously.
Once you have a good view of the droplets in the
Experiment O
Microwave Optics
Aims:
1. To investigate some properties of EM waves, such as refraction, polarisation and diffraction.
Objectives:
1. To gain experience planning, evaluating and refining experimental methods;
2. To develop experimental judgement in choosing the most suitable data collection methods;
3. To develop understanding of wave phenomena at microwave frequencies and relate this to
effects at visible frequencies.
3
4
Figure O.1: Some of the experimental apparatus. (1) Transmitter; (2) detector; (3) hollow prism; (4)
component mount; (5) goniometer; (6) partial mirror; (7) mirror and (8) polyethylene panel.
74
75
The slit on the transmitter produces a polarised
beam, and both the transmitter and receiver can
Safety Note Microwave Sources:
be rotated in their mounts to vary the angle of polarisation (the dial to rotate them can be seen in
The frequency of the microwaves has been
Figure R2.1). The receiver has a variable sensitivchosen such that they should provide no risk
ity scale. The scale reading should be multiplied
to the human body, however the manufacturby the factor indicated on the sensitivity scale to
ers caution that users should not look directly
obtain a true voltage reading.
down the beam from the transmitter.
Note that although some of the information
about these experiments is publicly available, the
material released by the manufacturers contains numerous errors which have been brought to the atIntroduction
tention of our staff. In addition to this, we have
made adjustments to various pieces of equipment
In this experiment you will be investigating some which mean that any results listed by the manuproperties of electromagnetic waves which are com- facturers may be invalid.
mon to all frequencies. The advantage of using microwaves over visible light is that small-scale effects become observable without the need for mi- 1 - Calculating the wavelength
croscopes etc.
You will have two lab sessions (8 hours of lab
time) to complete this experiment. Your assessment will be focused heavily on your experimental
notes and organisation, including planning, continuous review of results and refinement to experimental procedures. You should discuss the results in
each section fully before moving on to the next section. If sections 1 - 3 are complete you can still
achieve good marks even if section 4 is unfinished.
of the microwaves
Before you attempt this experiment you should

familiarise yourself with some classic experiments
which can be used to measure the wavelength of
electromagnetic waves. These should include (but
are not limited to) the Michelson-Morley interferometer, Fabry-Perot interferometers, Lloyds mirror
and double-slit interference.
Assignment
Experimental Apparatus
You are provided with:
One microwave transmitter
One microwave receiver
One goniometer with a fixed arm, a rotating
arm, and an angle scale
One rotating table (made from ethafoam)
Use two different methods to find the wavelength of

the microwaves, then compare the two values and
their errors. Take care when describing your experimental theory and procedures, and ensure you
design your experiments in order to find accurate
results.
You are welcome - encouraged even! - to use
methods not listed above.
2 - Refractive index of a prism
You have been provided with a hollow prism into

which you can place styrene pellets. The mould
One hollow prism (made from ethafoam)
should have a refractive index of 1 at microwave
frequencies.
Styrene pellets
By observing the transmitted signal as a function
of angle, calculate the refractive index of the
One polyethylene panel (for measuring Brewstyrene at the frequency of the microwaves. Consters angle)
sider how to ensure that the pattern you detect is
One simulated crystal (for Bragg diffraction solely a result of the styrene, not the hollow prism.
Discuss the features of the transmitted signal,
experiments)
and compare the refractive index at microwave freVarious full reflectors, partial reflectors and quencies to that of glass or silica prisms employed
in the visible part of the spectrum. Would you expolarisers
pect the refractive index to change if the styrene
Various component mounts
was a solid block, rather than pellets?
76
EXPERIMENT O.
1)ANoAelectricAfield
2)AAnAincidentA
electricAfieldA(blackA
arrowAindicatesA
directionAofAfield)
TheAelectronAcloudAisA
centredAonAtheAnucleus.
+A+
+
MICROWAVE OPTICS
+A+
+
3)AAnAoscillatingAelectricA
fieldAinducesAanAoscillatingA
chargeAseparationAonAtheA
atom,AwhichAcanAnowAbeA
consideredAanAelectricA
dipole.
Here,AtheAorangeAarrowA
indicatesAtheAdirectionAofA
propagationAofAaAphoton,A
andAtheAblackAcurveAshowsA
theAoscillationAofAtheA
electricAfieldAcomponent.
TheAelectricAfieldAinducesA
aAnetAseparationAofA
chargeAinAtheAatom.AAItAisA
slightlyAnegativeAonAtheA
right,AandAslightlyA
positiveAonAtheAleft.
Figure O.2: [Colour online] An electrically neutral atom (1) is exposed to a net electric field (2), which
induces a dipole. Oscillating fields (3) will induce an oscillating dipole.
3 - Brewsters Angle
The reflection of light from an interface can be understood as the combination of two processes - the
absorption of photons incident on the surface, then
the re-radiation by oscillating electric dipoles on
the surface.
Electric dipoles are simply a region of negative
charge close to a region of positive charge. They
can be permanent (e.g. a H2 O molecule in which
the oxygen atom has a net negative charge and the
hydrogen atoms have a net positive charge; the
molecule is still electrically neutral, but will respond on a small scale to electric fields) or induced
and temporary. Figure O.2 shows a sketch of an
EM wave inducing a dipole.
however dipoles are unable to radiate photons in a

direction parallel to their axis of oscillation.
When light is incident on an interface we normally consider two polarisations: s-polarised,
where the electric field is oscillating in a plane parallel to the interface, and p-polarised, where the
electric field is oscillating perpendicular to the interface.
You will be studying the reflection and transmission of both polarisations at an interface between
air and a polyethylene panel. The behaviour of the
two polarisations is different, and you will use this
to identify the refractive index of the panel, and
characterise the transmitter.
Assignment
When electromagnetic waves are incident on an

electric dipole, the oscillating electric field causes The transmitter has a slit on the front face, which
the dipoles to oscillate in the same direction. The will be either horizontal or vertical when the angle
photons are absorbed and then quickly re-radiated, is set to zero. It is not known which slit direction
77
corresponds to s- or p-polarisation. Set the transmitter to output equal intensities of horizontally
and vertically polarized light - this should be at
approximately 45o but it may not be exact.
Position your polyethylene panel on the rotating
table in the centre of the goniometer, and collect
both the reflected and transmitted signals in both
polarisations (i.e. with receiver set to 0o and 90o )
as a function of incident angle. Consider carefully
where to position the receiver as the panel rotates.
Figure O.3 shows a sketch of p-polarised light
incident on an interface between two media with
different refractive indices (in this case, n2 > n1 ).
The light induces electric dipoles on the interface.
For p-polarised light the dipoles oscillate parallel to
the blue arrow on the incident wave. Brewsters
angle occurs when the reflected ray would also be
parallel to this direction of oscillation. It occurs
because electric dipoles are unable to emit radiation along their oscillation axis - they cannot emit
photons in that direction, so the intensity of the
reflected ray drops to zero.
Incident
wave
Reflected
wave
1
n1
n2
2
Transmitted
wave
Figure O.3: [Colour online] Light incident on an

interface between media with refractive indices n1
and n2 at angle 1 is partially reflected and partially transmitted. The transmitted beam is refracted to angle 2 . The blue arrows show the direction of oscillation of the electric field - in this
case, all three beams are p-polarised.
Use your collected data to identify Brewsters angle, the angle at which the p-polarised component
of the beam has zero reflected intensity, and is instead completely transmitted.
The refractive index of the panel, np , can be
found from the relationship between the refractive
index of the air na and Brewsters angle, B .
tan B =
np
na
(O.1)
You can derive this relationship by considering

Snells law and the angles in the system.
Once you have identified Brewsters angle, calculate the refractive index of the panel. When the slit
is vertical, does is correspond to p- or s-polarised
light? Why doesnt s-polarised light exhibit an
equivalent effect at Brewsters angle?
4 - Total internal reflection

(TIR) & evanescent waves
The theory for this section is complex and you may
not understand it mathematically in your first year,
however the reasons behind the effect can be understood. This experiment also demonstrates an
effect very similar to quantum tunnelling, so the
visual example may be of use when you encounter
the theory later in your physics career.
Theory
When light is totally internally reflected its momentum in the plane of the interface remains constant
(e.g. In Figure O.3 the reflected wave is still travelling from left to right), however its momentum
perpendicular to the interface is reversed. Therefore, although the energy of the beam should remain constant, conservation of momentum appears
to have been violated.
The solution is an evanescent wave - a wave
which does not transport any energy but decays
in amplitude as its distance from the interface
is increased. These waves have a purely imaginary momentum; when the total momentum before and after striking the interface is calculated,
|p|2 = p2x + p2y + p2z , the inclusion of the imaginary
component means that momentum can be seen to
be conserved.
These waves can be observed when a second
prism is introduced into the field of the evanescent
waves. The prism has a high enough refractive index that if there were no air gap the light would
not experience TIR and some would be transmitted. As the air gap increases the light is forced
first to convert from its radiative (normal) mode
into evanescent waves, but then it reaches the second prism and converts back to normal light and
can be detected. The energy has tunnelled across
the air gap. As the air gap increases in width, the
amplitude of the evanescent wave should decay, and
this should be observed as a corresponding decrease
in the intensity of the transmitted waves.
This effect can be described mathematically in
the same way as quantum tunnelling, and it has
helped develop many current areas of research, such
as quantum computing and microwave frequency
invisibility cloaks.
78
EXPERIMENT O.
MICROWAVE OPTICS
Detector
Prism 1
Transmitter
(fixed)
Incident
beam
Reflected
beam
Separation, d
Transmitted
beam
Prism 2
Figure O.4: Incident wave is totally internally reflected (p is greater than the critical angle). Reflected
and transmitted beams are shown, and the separation between the prisms is d.
Experiment
Set up your system so that your beam from your
transmitter is totally internally reflected within the
prism. Once you have achieved TIR, keep your
prism stationary and plot the intensity of the reflected beam as a function of receiver angle to observe the spread of the signal.
Without moving your first prism, bring another
wax prism into contact with the face of the first
prism which is reflecting the beam and measure the
change in reflected intensity and any transmitted
signal as a function of angle.
Observe the change to the reflected and transmitted signals when the second prism is gradually
moved away from the first prism (i.e. introducing
an air gap which gradually increases in width). Try
to keep the two sides of the air gap parallel. The
setup is shown in Figure O.4. You may need to adjust the positions of the detector for the reflected
and transmitted beams slightly to find the highest
signal at each separation.
Plot a graph of transmitted intensity vs prism
separation. Does this decay exponentially? What
is the approximate distance over which the signal
changes by 50%? How does this compare to the
wavelength of the microwaves? The physics is also
valid for visible light - discuss the implications of
trying to observe it.
Experiment Q1
Photoemission of Electrons from

Metallic Surfaces
Aims:
1. To plot graphs of anode potential against photoelectric current for all emission lines of a
mercury source;
2. To determine the stopping voltage for those frequencies corresponding to the spectral lines;
3. To determine Plancks constant.
Objectives:
1. To learn to set up and calibrate an optical apparatus;
2. To collect data in an intelligent way, e.g., to record more values close to the stopping voltage.
Hg Lamp
Prism
S1
L1
L2
S2 +
Photocell
Oscilloscope
Photoemission
amplifier
Figure Q1.1: The apparatus for studying the photoelectric effect.
Further reading:
Company, New York, 2004). Chapter 34.
G.P. Harnwell, Principles of Electricity and Electromagnetism 2nd Edn. (McGraw-Hill, New
York, 1949). pp 215220.
M. Sayer & A. Mansingh, Measurement, Instrumentation and Experiment Design in Physics
and Engineering (Prentice-Hall India, New Delhi, 2000). pp 116120.
79
80
EXPERIMENT Q1.
PHOTOEMISSION OF ELECTRONS FROM METALLIC SURFACES
Introduction
Electrons may be ejected from an electrical conductor by electromagnetic radiation incident on its
surface. This is known as photoemission.
Aside from its practical applications, the photoelectric effect is of huge importance as one of the
best-established foundations of quantum theory. It
is found that electrons are emitted only if the frequency, , of the radiation exceeds a threshold
frequency, 0 , that is characteristic of the surface.
For most metals, 0 is in the visible or near ultraviolet region of the spectrum. The greatest kinetic
energy, E, of the emitted electrons is given by:
E = h( 0 )
(Q1.1)
estimate of the work function (A ) of the anode in

the photocell should also be made.
Light from a mercury vapour lamp is split into its
constituent spectral lines. The individual lines are
directed onto a photocathode (C), which is a thin
layer of potassium on one wall of an evacuated glass
vessel. It has been specially prepared to reduce its
work function below that of pure potassium. A
second electrode within the vessel, the anode (A),
consists of a loop of platinum wire parallel to C and
a few millimetres from it.
Electrical connections are made to C and both
ends of A. When electrons emitted from C reach A
they contribute to the current in an external circuit. To determine the maximum kinetic energy of
these electrons, A is made progressively more negative with respect to C until the photocurrent in the
external circuit just reaches zero. We then have:
where h is Plancks constant.

Classical electromagnetic theory fails to explain
this relation between energy and frequency, or the
E = eV0
(Q1.3)
fact that the number of emitted electrons is proportional to the intensity of the radiation. Ein- where V0 is the potential difference (p.d.) between
stein received the Nobel Prize for his explanation the surfaces of C and A for zero photocurrent and
of the experimental results. This explanation can e is the charge on an electron (1.60 1019 C).
be summarized as follows :
Two factors complicate the measurements:
1. Electromagnetic radiation consists of quanta
(photons) each with energy h.
2. A photon absorbed by the surface may impart
its energy to one or more electrons.
3. The probability that an electron acquires energy from more than one photon is negligible.
If the least energy needed to free an electron from
the surface is (the work function), the greatest
kinetic energy of the emitted electrons is:
E = h
(Q1.2)
The experimental set up used here is essentially

that of Lenard (1902) and others. Millikan devised
an improved method and was awarded the Nobel
Prize in 1924 for his experimental work in this field.
With care and good experimental technique this
apparatus will give good results. You must ensure that you have the optimum alignment for each
wavelength and decide on the optimum source slit
size to use. Take care to make a thorough error
analysis of your results.
The Experiment
In this experiment Plancks constant (h) is to be
determined from measurements of the energy of
electrons emitted from the surface of a potassium
photocathode by light of known frequencies. An
(1) The p.d. applied externally between A and C

differs from that between their surfaces by the
contact potentials of the materials from which
they are constructed (see Harnwell).
Since the contact potentials are constant, for
constant surface conditions, this problem is
overcome by observing differences in external
p.d. for zero photocurrent with differences in
the frequency of the light reaching C.
(2) The photocurrent to be observed is not the
only source of current in the external circuit.
Other significant contributions are:
(a) photoemission from C by scattered radiation at unwanted frequencies;
(b) photoemission from A by ultra-violet radiation, and from contaminants on A by
visible and ultra-violet radiation;
(c) conduction through, and along the surface of, the glass envelope of the photocell.
The system was designed to minimise the effects of unwanted currents but they cannot be
eliminated entirely. They can be distinguished
from the required photocurrent because they
vary slowly as the p.d. between A and C is
varied. Thus, as A is made progressively less
negative with respect to C, the appearance of
the required photocurrent is seen as a change
in slope of the I V curve (see Figure Q1.2).
81
The Optical System
photoelectric
current, I
saturation
current
Important: The cathode of the photocell

must be protected from excess illumination at
all times, as this will cause temporary (sometimes permanent) loss of sensitivity. Do not
allow full daylight or the undispersed light
from the lamp to reach the photocathode.
anode potential
w.r.t. cathode, V
stopping
potential
Figure Q1.2: Current vs. p.d. in the photocell.
A Note on Contact Potential

The energy absorbed from a photon by a conduction electron at the surface of a metal enables it to
escape from the metal. The energy needed to cross
the surface is eC , where C is the work function of
the cathode in volts. However, the potential difference between the surfaces of the cathode and anode
is not the voltage, Vm , measured by the meter. In
fact, Vm represents the potential difference between
F
F
, of the two metal surand EA
the Fermi levels, EC
faces. (The Fermi level of a metal indicates the
highest occupied electron energy state of the metal
at 0 K). From Figure Q1.3, it can be seen that:
V = Vm (C A )
(Q1.4)
eV
h
Note that careful alignment of the optical system is crucial to maximise the ratio of wanted to
unwanted light reaching the photocathode.
A plan view of the optical system is shown in
Figure Q1.4. The lens L1 should focus an image
of the mercury discharge tube on to slit S1 . The
width of S1 can be varied to control the intensity of
light reaching the photocell and hence the magnitude of the photocurrent. The lens L2 should focus
an image of S1 on to the card carrying S2 .
The prism is used to separate the spectral lines.
It must be rotated to obtain the best resolution of
each spectral line as required. Do not make the
spectrum too wide keep it below 10 cm. The
wavelengths in nanometres of the strongest spectral lines from the mercury lamp that are relevant
to this experiment are given in Table 1.
Wavelength (nm)
Colour
578
Yellow
546
Green
436
Blueviolet
405
Deep violet
365
Ultraviolet
e A
Table Q1.1: The main emission lines of mercury.
eC
EA
Note: the eye is insensitive to the 365 nm line but

the card around S2 (like most white paper) contains
a fluorescent compound that emits blue light when
excited by light of shorter wavelength. The detector carrying S2 can be moved across the bench to
Cathode
Vacuum
Anode
select the desired spectral line. N.B. The glass envelope of the mercury lamp filters ultra-violet light
Figure Q1.3: The electron energy levels in the pho- so the UV line will be rather weak.
toelectric cell.
E FC
eVm
The Electrical System

The contact potential is the difference in the work
functions of the metal surfaces (the term arises
from studies of junctions between pairs of metals).
For this experiment you just need to know that a
contact potential exists. A full understanding of
Fermi Levels in metals will have to wait until the
second and third year condensed matter courses.
A simplified block diagram of the electronics unit

is shown in Appendix D. You do not need to understand the detailed operation of this circuit.
The cathode of the photocell connects directly
to the input of an amplifier (a current-to-voltage
converter) mounted on the photocell itself. The
82
EXPERIMENT Q1.
Mercury
Lamp
L1
S1
L2
Prism
S2
Phot
ocell
Figure Q1.4: The optical layout of the photoelectric effect apparatus.
cathode photocurrent is converted to a voltage and

an oscilloscope is connected to the detector unit enabling you to visualise the output from the photocell. You will see that if the brightness is too high,
the waveform is clipped as the photocell-mounted
amplifier is overloaded. You may find this visual
indicator useful in adjusting the slit width.
Since the mercury lamp is powered from the
50 Hz mains supply, its light output will fluctuate
at 100 Hz, as will the desired component of the photocurrent. Because of this, the lamp and the electronics unit must be connected to the same wall
outlet (details in Appendix D).
The photocell anode-to-cathode potential is controlled by the front panel control marked Anode
Bias and is displayed on the right hand digital
panel meter (DPM). It has a range of roughly +1
to 2.5 V. The other DPM gives a reading in picoAmp`eres of the average photocurrent.
Cleaning the Anode
Method
In this experiment a photocell is illuminated with
light of different wavelengths. The stopping voltage, i.e. the voltage required to eliminate any photocurrent, is determined in each case, and by combining the stopping voltages for all the wavelengths
of illumination, a value for the work-function of the
platinum anode in the cell is determined.
There are several points to be aware of while performing the measurements:
(a) The photocell is sensitive to excessive amounts
of light it should be kept covered when not
actually being used.
(b) Precise values of the stopping voltage are hard
to obtain when the photocurrent is low. Try
to maximise the photocurrent by optimising
the lensprismslitdetector assembly for each
wavelength in turn - aim for at least 100 pA at
the highest voltage setting when illuminating
the photocell with each line.
A major contribution to unwanted photocurrent is

emission from the anode. The work function of a (c) You are interested in determining the potenclean platinum surface is sufficiently high to pretial at which the photocurrent initially turns
vent photoemission, except when it is excited by
upwards from the zero level. The zero level,
far ultra-violet light. However, any contamination
may actually be offset from zero by a signifof the surface will lower its work function. The best
icant amount. You should measure sufficient
available means of removing contamination from
points that you can accurately determine both
the surface of the anode is to heat it.
the zero level and the initial upturn in phoIn this apparatus the anode can be heated by
tocurrent with decreasing potential difference.
passing a current through it. The push button on
the PSA unit marked Purge Anode performs this
function. The heating period is timed and after it Data Analysis
has elapsed, a lockout period follows during which
it is impossible to re-heat the anode (to prevent it The data analysis method is detailed below.
from overheating).
If the photocell is illuminated with blue light and (1) Determine the errors in the bias voltages and
the anode potential is taken to its maximum negphotocurrent readings.
ative value, the reading on the photocurrent meter will be a measure of anode surface contamina- (2) For each wavelength from the mercury lamp,
tion. The anode surface will be as clean as can
determine the saturation photo-current. These
be achieved when repeated bursts of heating fail to
figures will be useful later in deciding the relreduce this reading.
ative accuracy of each stopping voltage.
83
(3) For each wavelength, plot a graph of photocurrent versus bias voltage. In each case:
Appendix D
The Electrical Circuit
(a) Fit a suitable straight line to the region of

your data with large negative bias i.e. the The electrical circuit diagram is shown in Figure
region in which there should be no photo- Q1.5.
electrons excited from the photocathode;
The cathode of the photocell connects directly to
(b) Extrapolate this straight line across the the input of an amplifier. This converts the cathode
whole range of bias voltages;
photocurrent to a voltage which is displayed on the
(c) Identify the point at which the photocur- detector unit and an oscilloscope.
This arrangement ensures that the cathode is
rent begins to increase from this baseline. This point will give you a measure- maintained at zero volts, because it is connected
ment of the stopping voltage, although to the virtual earth point of the amplifier (the ve
input of the op-amp is referred to as a virtual earth,
further corrections will be needed;
because the p.d. between the ve and +ve inputs
(d) Use your knowledge of the precision with
is negligible.).
which the photocurrent can be measured
Since the mercury lamp is powered from the
and the sensitivity of the photocurrent to
50
Hz mains supply, its light output will fluctuate
changes in the bias voltage to estimate
at
100
Hz, as will the desired component of the phothe accuracy of the measured stopping
tocurrent.
The phase sensitive amplifier (PSA) will
voltage for each wavelength.
reject steady leakage current in the photocell and
The metered bias voltage Vm differs from the ac- amplifier, as well as stray daylight reaching the photual voltage, V , by an amount related to the work tocell; it will only respond to light fluctuating at a
specified frequency. The timing of the PSA (which
functions of the anode and cathode:
ensures its response is to the mercury lamp only)
Vm = V + (C A )
(Q1.5) is also derived from the 50 Hz mains supply. The
PSA will only be synchronized to the mercury lamp
The stopping voltage V0 is that voltage at which if they are both supplied by the same phase of the
the kinetic energy of the photoelectrons is just remains supply. To ensure that this is the case, it is
duced to zero, i.e. the energy eV0 is equal to the imperative that the lamp and the electronics unit
difference between the energy of the photons and are both connected to the same wall outlet. For
work function of the cathode:
further explanation of phase sensitive detection, see
Experimental Measurement by Sayer and ManseV0 = h eC
(Q1.6) ingh.
However, the actual stopping voltage is the metered
value, Vm,0 less (C A ), so that:
e [Vm,0 (C A )] = h eC
and:
Vm,0 =
h
A
e
(Q1.7)
(Q1.8)
(5) Plot a graph of Vm,0 against frequency, .

(6) Determine the gradient and intercept either
by using Origin, or drawing plausible straight
lines with maximum and minimum likely
slopes through the data points.
(7) From the resulting straight-line equation, determine a value for Plancks constant, h, and
its error, and a value for the work function of
the platinum anode, A , and its error.
(8) Comment on the biggest sources of error in the
experiment. How might they be reduced?
84
EXPERIMENT Q1.
Cathode
I to V
converter
Anode
Phase sensitive
amplifier (PSA)
+1V
purger
~1V
DPM = Digital Panel Meter
Bias
setting
2.5V
DPM
DPM
Bias
Voltage
Photocurrent
Figure Q1.5: A simplified circuit diagram for the detector electronics in .
Experiment Q2
Charge:Mass Ratio of the Electron

Aims:
1. To use the effect of electric and magnetic fields on a stream of electrons in order to determine
the ratio of the electronic charge to mass.
Objectives:
1. To understand helical paths of electrons in a magnetic field and focusing at integral turns
of the helix.
2. To complement the Level 4 Physics lectures on Fields.
Figure Q2.1: The e/m apparatus
Further reading:
Company, New York, 2004). Chapters 21-6, 26-1 & 26-2.
Safety Note:
This experiment uses high voltages. Ensure the 1000 V power supply is switched off whenever
you connect up the circuit or modify it. Do not touch any part of the circuit when the power
supply is switched on. The power supply is protected so that it cannot deliver enough current to
be lethal, but an electric shock might cause you to damage yourself or the apparatus.
85
86
EXPERIMENT Q2.
CHARGE:MASS RATIO OF THE ELECTRON
Introduction
alternating potential difference between either of

two pairs of deflector plates, X and Y.
The force on an electron with charge e and mass
Also provided is a solenoid which connects to the
m, moving with velocity v in a magnetic field B 0 - 5A current supply, to produce a magnetic field
and an electric field E, is
which may be varied. The solenoid is deceptively
heavy and must be placed carefully around the
F = e(E + v B)
cathode-ray tube to avoid damage. The solenoids
will heat up during use, so please turn off the curwhere F is measured in Newtons (N), e in Coulombs
rent supply when it is not needed. The number of
(C), E in V/m, v in m/s and B in Tesla (T). The
turns indicated on the solenoid is the total number
acceleration of the electron due to this force is:
of turns, not the number per unit length.
a = (E + v B).(e/m)
(Q2.1)
and therefore measurements of the electrons motion in known magnetic and electric fields can be
used to find the charge:mass ratio of the electron,
e/m. Note that neither e nor m can be found separately from a measurement of this type as only
their ratio appears in the equation of motion.
J. J. Thomson is credited for the discovery of the
electron in 1897. Prior to this, the hydrogen atom
was believed to be the lightest particle in existence.
Thomson devised an experiment to measure the deflection of charged particles in a magnetic field perpendicular to the direction of motion of the particles. From this he calculated values of the charge to
mass ratio and showed unequivocally that the electron (then known as a cathode-ray) was nearly
2000 times lighter than the hydrogen ion.
The experiment you will perform today is an ingenious adaptation of Thomsons, first developed
by H. Busch in 1922. The charge:mass ratio of electrons is to be calculated from measurements of the
cyclotron, or helical, motion of electrons subjected
to a parallel magnetic field in a cathode-ray tube.
Theory
The ratio e/m can be found from the path taken by

the electrons in passing through a region of known
electric and magnetic fields.
An electric field alone will not suffice (unless an
independent measurement of the electron velocity
is available) because the path in a purely electric
field is independent of e/m (see Appendix E).
Where there is a magnetic field as well, however,
the acceleration has a component proportional to
v; the deflection then depends on (e/m)1/2 . Hence,
e/m can be found from the path of the electron
through the field.
The electric and magnetic fields may be arranged
in various ways to measure e/m. An obvious
method, with the cathode-ray tube available, would
be to apply a magnetic field transverse to the electron beam and to calculate e/m from the deflection of the spot across the screen. It is, however,
difficult to achieve much accuracy by this method
unless a specially designed CRT is used; otherwise,
the position of the spot is ill-defined and it is not
easy to measure (or calculate) the magnetic field in
Apparatus
the region traversed by the electrons.
These difficulties are avoided in this experiment
Figure Q2.2 shows a diagram of the cathode-ray
by
using the focusing effect of a longitudinal magtube (CRT) and the associated power supply cirnetic
field on the nearly parallel beam of electrons.
cuits. Electrons are emitted by the indirectlyheated cathode, K, and are accelerated towards the
anode, A, whose potential is 300 1000 volts more Motion in a Magnetic Field
positive than K. The anode is earthed to the metal
of the case so in this case, the cathode is the dan- Consider the electron beam as it emerges from the
small hole in the last disc of the anode. Since all
gerous electrode to touch.
The electric fields in the anode-cathode space are the electrons start from the cathode with zero vearranged so that the electrons are focused into a locity and are accelerated through a rise in electric
narrow beam which passes through a small hole potential V (applied between the anode and cathin the anode and eventually strikes the fluorescent ode), they all have the same velocity v, given by:
screen at the end of the tube. The number of elec1
trons in the beam, and hence the brightness of the
mv 2 = eV
(Q2.2)
2
luminous spot on the screen, is controlled by the potential of the electrode G (usually called the grid). The directions of motion, however, are slightly
After leaving the anode, the beam may be de- spread around the axial direction of the tube, beflected from the axis of the tube by applying an cause the cathode has a finite area.
87
Fluorescent Screen
Deflector
Plates "X"
Deflector
Plates "Y"
(Red) (Red)
Anode A
Control Box
+ Circuits (Blue Box)
Control
Electrode G
Cathode K
3 4 62
5 71
Heater
+
470k
10M
To Power
Supply
Unit
500k
Brightness Control
250k VG
6.3V
a.c.
Figure Q2.2: The cathode-ray tube and supply circuits for the measurement of e/m.
88
EXPERIMENT Q2.
Suppose that an electron leaves the aperture in

the anode with a small but finite component v of
velocity perpendicular to the axis. The component
vz parallel to the axis is almost the same as v, if
v v. In a uniform magnetic field, B parallel
to the axis, the force e(v B) experienced by the
electron has magnitude ev B and is perpendicular
to the axis and to v . Its effect is to rotate the
direction of v , but since it is always normal to it,
to leave its magnitude unchanged. The resulting
motion is circular (see Appendix F) and the rate
at which the direction of v rotates is given by:
e
B
=
m
This is constant for a given B.

which is obtained by substituting from equations
(Q2.2) and (Q2.3). Rearranging, we obtain the following expression for e/m:
8 2 n2 V
e
= 2 2
m
l B
(Q2.5)
Thus it is possible to determine e/m by finding the

various combinations of fields for which the beam
is focused; this forms the basis of the experiment.
The Field Inside a Solenoid

The field inside the solenoid is given, as a first approximation, by the formula:
B = 0 N I
Helical motion
Since the electron already has a large component
of velocity along the axis of the CRT (induced by
the accelerating electric field), the circular motion
is drawn out into a helical path towards the screen,
rather than a straight line as it would be in the
absence of B. The radius of the helix is:
v
v . e
r=
=
B m
and the period of the circular motion is:
2 . e
2
=
(Q2.3)
T =
B m
Note that T depends only on B and e/m, not upon
the velocity of the electron. This is the basis of
the action of the cyclotron, in which electrons are
accelerated to high speeds as they revolve in a magnetic field. This effect is also of importance for the
behaviour of electrons in solids.
where N is the number of turns per metre of its

length and I is the current in Amps. This formula
would be exact if the solenoid were infinitely long.
A correction is necessary because its length is finite.
Z
Y
X
V
P
A
D
Figure Q2.3: A cross section through the solenoid,

showing the lengths in eq. Q2.7.
Figure Q2.3 shows a cross section through the

solenoid, indicating various lengths associated with
an arbitrary point P on the axis of the solenoid.
The magnetic field varies along the axis of the finite
Magnetic Focusing
solenoid; for the point P, a distance X from the
end,
it can be shown that the true value of B is:
The period is the same for all electrons, therefore it
is possible to choose B so that all electrons make an
integral number (n) of rotations as they travel from
B = 0 N I.K(X)
(Q2.6)
anode to screen. In this case, since they all emerge
where:
from a small hole in the anode, they all arrive at

the screen within a similar small area and produce a
X
A+D+Z
K(X)
=
log
+
sharply focused spot. An odd half-integral number
e
2D
A+Y
of rotations, on the other hand, brings the electrons

U
A+D+W
to their furthest from the axis when they strike the
loge
(Q2.7)
2D
A+V
screen and the spot is badly focused.
If the anode-screen distance is l, the transit time
and the lengths A, D and U Z are shown in Fig.
of the electrons is:
Q2.3.
t = nT = l/vz
To help with the calculations, there is an Origin file provided within the Q2 folder on the DLM
and the number of rotations, n, is:
which will calculate values of K automatically
l
l
when you enter the various solenoid dimensions.
'
n =
T vz
Tv
The mean B field encountered by the electrons in
r
their passage from the anode to the screen should
lB e
1 m
=
(Q2.4) be used in calculating e/m.
2 m
2V e
89
Method
Important
Damage to CRTs: The e/m experiment
involves focusing an electron beam onto a
phosphor screen, which is easily burnt by the
beams. Keep the beam brightness to a minimum at all times and ensure that the apparatus is switched off when not in use.
Damage to solenoids: When a current flows
through the solenoid, it quite naturally leads
to heating of the coil. Avoid leaving this
switched on for too long, as it can cause the
insulation to melt.
1. Briefly investigate the effects of varying the potentials of the different electrodes and confirm
that the deflection of the beam is in the sense
expected for negatively-charged particles travelling from the cathode to the screen.
2. Sketch the paths of two electrons that have
velocity v parallel to the magnetic field B and
their transverse components of velocity, v , in
opposite directions; first draw the paths for
one complete cycle and then for half a cycle.
What would you expect to see on the screen?
3. Now measure e/m by finding the various fields
for which the beam is focused. With zero voltage applied to the XY deflector plates and
several fixed values of accelerating voltage (between 700 and 1000 V), vary the magnetic field
and find sets of foci for consecutive values of
n. An axial magnetic field may be produced
by the solenoid provided, using the d.c. power
supply. The field may be calculated from the
measured current and the dimensions of the
coil.
to a short curly squiggle rather than a spot discuss whether the use of the alternating voltage is
beneficial to your experiment.
Points to note
There are several points to note whilst performing
the measurements:
The EHT voltage may fluctuate, or show longterm drift, throughout the duration of the experiment. You should be aware of this and
compensate for it where possible.
Large beam currents, which occur at higher
accelerating voltages, will tend to overload
the EHT supply, resulting in a drop in the
recorded value of accelerating voltage as you
turn up the brightness. Therefore, it is a good
idea to avoid adjusting the brightness control
once you have chosen a value for V . For each
value of V , you should set the brightness to
the lowest usable setting (see next point) before varying B.
It is comparatively easy to burn a hole in the
phosphor of the cathode-ray tube (CRT), by
leaving for prolonged periods with a focused
spot showing on it. You should avoid this at
all costs, as it makes your (and everyone elses)
measurements more difficult since the spot becomes hard to see, and the CRTs are expensive
to replace and becoming hard to obtain.
Without a magnetic field applied, the focus is
created by the electron beam optics (i.e. electrostatic focusing via a set of electrodes) and is
not very sharp. Therefore, the first magnetic
field induced focus tends to require a larger B
field increment than subsequent ones.
In addition to the currents and voltages discussed above, you will need to measure the:
distance l, which should be taken from

The values of B for the best focus should be multhe centre of the four focusing electrodes
tiples of the value for n = 1. If they are not, it is
in the CRT, to the screen;
likely that there is some focusing of the electron
dimensions of the solenoid, i.e. its length,
beam before it emerges from the anode. If this
L, diameter and thickness, D;
happens, the length l should be measured correctly
position of the CRT within the solenoid.
from the point where the diameter of the beam is
least, which is not necessarily at the final anode
and may vary with the magnetic field and with the
Data Analysis
accelerating potential. Experience suggests that l
is best measured from the centre of the set of 4
The data analysis methods are detailed below.
anodes.
An alternating voltage may be applied to the (1) Estimate the error in your reading of the EHT
plates so that the spot appears drawn out into a
voltage, V . You can assume that the digital
line (a suitable 13 V a.c. supply is provided). The
display has been calibrated. Discuss the causes
collapse of the line into a spot indicates that the
of any variation in the voltage, and what you
deflection has vanished. Often the line collapses
might do to minimise their effect.
90
EXPERIMENT Q2.
(2) Similarly, estimate the error in reading the

solenoid current.
(3) The solenoid current also depends on the precise focal point selected.

(8) While it is possible to work with a single value
of V , it is possible that there could be a systematic error in it. This error can be avoided
2
by plotting (g(V )) against V . This ought to
pass through the origin, any systematic error
will show up as a non-zero intercept.
(a) Discuss the difficulties involved in finding

a sharp focus.
(9) Determine the slope of this graph, G say:
(b) By repeating one set of measurements
several times estimate the actual error in
8 2 me
(Q2.10)
G=
determining the correct solenoid current
el2
for a given n i.e. estimate the reproand its error.
ducibility of your results.
(c) Is your error constant with n?
(10) Writing the ratio e/me as q, we can find the
value of q and its error by rearranging the
(4) Is there any error in n itself?
above:
(5) The B field is a linear function of the solenoid
current. You need to estimate a value for
8 2
q
=
(Q2.11)
the correction factor K, arising because the
Gl2
solenoid is non-ideal. To do this:
(11) Discuss the largest source of error in your
(a) consider the dimensions of the solenoid,
value. How might you reduce this error?
and your estimates of their errors;
(b) use the Origin Calculator provided for K,
and estimate the size of the error in K
that might arise from the uncertainties in
these dimensions;
(c) determine an average value for K, taken
over the length of solenoid actually involved in the magnetic focussing.
(d) estimate a final error in K, combine it
with your error for the solenoid current I,
and determine a value for the error in B,
using the relationship you found in your
pre-lab assignment.
Consider the relationship between n, B and V :
r
8 2 me V
(Q2.8)
B=n
el2
For a given value of V , there is a linear relationship between n and B. In principle, we might
expect this line to pass through the origin. However, the n = 0 point is not valid, for the reasons
mentioned above, so, instead we rely only on the
gradient g which will vary with V :
r
dB
8 2 me V
g(V ) =
=
(Q2.9)
dn
el2
(6) For each value of V , plot a graph of B against
n, clearly indicating your error bars.
(7) In each case, determine the gradient, g(V ),
and the error in g(V ). You can determine the
error by either (a) drawing plausible straight
lines through each set of points, to find the
maximum and minimum likely values, or (b)
using the linear fit method in Origin.
91
Appendix E
Appendix F
To show that the deflection of a charged particle under the influence of an electric field,
alone, is independent of e/m:
It is not difficult to see why this should be. Consider an electron travelling at velocity vz parallel to
the z-axis. Assuming that the electron acquired its
velocity from an accelerating voltage, V , we have:
To demonstrate that motion in a perpendicular magnetic field is circular:

We can analyse the motion as follows, considering
only the component of velocity perpendicular to
the magnetic field, v . Consider the components
of v parallel to the x and y-axes. The transverse
accelerations will be given by:
1
mv 2 = eV
2 z
or, rearranging:
ax =
e
dvx
= vy B
dt
m
(Q2.13)
e
dvy
= vx B
(Q2.14)
dt
m
(Q2.12) If we differentiate equation (Q2.13) and substitute
from equation (Q2.14), we obtain:
A transverse electric field Ex is now applied parallel
to the electron. In the time t = z/vz that the
d2 vx e 2 2
+
B vx = 0
electron takes to traverse a short length z of its
dt2
m
path, the electron acquires a transverse velocity vx ,
which is an equation for simple harmonic motion
and its direction of motion changes by an angle:
(SHM). A similar equation can also be found for
vy . The two equations may be solved together, and
' tan = vx /vz
have solutions of the form:
where vx is given by:
vx = v cos t; vy = v sin t
vx = ax t
in which is the angular frequency of the oscillaand ax is the transverse acceleration. Substituting tory motion. Thus, vx and vy are 90 out of phase
for t gives:
with each other, and the motion is circular with
ax z
frequency given by the equation of SHM:
= 2
vz
e
B
=
From equations (Q2.1) and (Q2.12) it is clear that
m
2
both ax and vz are proportional to e/m, with the
result that:
= (Ex /2V )z
ay =
e
vz2 = 2V.
m
i.e. the deflection is independent of the charge:mass

ratio of the electron.
Experiment R1
Thermionic Emission from Tungsten

Aims:
1. To determine the cathode temperature of a vacuum tube diode by measuring its resistance
when various currents flow through it;
2. To check the validity of Richardsons Law;
3. To obtain an estimate of the work function of the cathode metal.
Objectives:
1. To gain experience of and appreciation for the importance of careful measurements;
2. To appreciate the distribution of thermal energies of electrons;
3. To realize the importance of the metal surface on the value of the work function.
Figure R1.1: The thermionic emission apparatus.
Further reading:
Bleaney & Bleaney, Electricity and Magnetism (1st edition), pp. 84-90 & 329-336.
92
93
Table R1.1: Work functions, , for various metals.

Safety Note:
This experiment uses high voltages. Ensure
the 150 V power supply is switched off whenever you wire up or modify the circuit, and
always use shrouded cables with the high voltage sockets. Do not touch any part of the circuit when it is switched on. An electric shock
might cause damage to you or the apparatus.
Introduction
At ordinary temperatures, the energy, E, of a conduction electron inside a metal is much less than the
energy, E0 , that it would have were it at rest and
far outside the metal. Unless the electron somehow
acquires an additional energy = E0 E, it cannot escape. The quantity of work, , required to
extract an electron from a metal when the temperature is low enough for the thermal contribution to
E to be negligible is termed the work function of
the metal. A few representative values, expressed
in electron-volts (eV), are given in Table R1.1.
At higher temperatures, some electrons share the
thermal vibrations of the crystal and acquire additional energies1 of the order kB T , where T is the
absolute temperature and kB is Boltzmanns constant (1.38 1023 J K1 , or 0.86 104 eV K1 ).
Even at temperatures as high as 3000 K, kB T
amounts to rather less than 0.3 eV, and the mean
energy of the electrons remains far below E0 . At
any time, however, there are a few electrons with
thermal energies much greater than the average. If
the total energy of one of them exceeds E0 , it is energetically possible for it to escape from the metal.
Whether it does so depends on its position and velocity relative to the surface. This is the basis of
thermionic emission of electrons, the phenomenon
behind the production of electrons in, for instance,
cathode ray tubes or X-ray generators.
The number of electrons whose thermal energies
exceed , and, therefore, also the number escaping
in unit time, may be shown by a statistical argument to be proportional to exp(/kB T ). A complete analysis leads to the result that for any metal
the number of electrons emitted per unit time from
unit area of the surface, n, should be:
Metal
(eV)
K
Na
Fe
Th
Zn
Cu
W
Pt
Th on W
2.2
2.3
3.2
3.3
3.4
4.5
4.5
5.3
approx. 2.7
The Experiment
Thermionic emission from the cathode of a vacuum
tube diode is to be studied by measuring the anodevoltage versus anode-current characteristics for various cathode temperatures. A diode is provided in
which the cathode is a thin metal wire, and the anode is a box-shaped structure surrounding it. The
cathode may be heated electrically to temperatures
in excess of 2000 K. At these temperatures electrons
are emitted thermionically in large numbers.
The behaviour of the electrons after emission depends on the electric field that they encounter between the cathode and anode. Three more or less
distinct situations may be defined (you should be
able to identify these regions in your graphs):
A. Anode potential negative with respect to cathode

The field between the electrodes opposes the motion of the electrons towards the anode. A few may
be emitted with sufficient velocity to reach it and
drive a current through the external circuit, but the
rest merely return to the cathode.
B. Anode slightly positive:

charge limited currents
space-
When the anode is at a higher potential than the

cathode, electrons accelerate towards it and a current flows through the diode. It might at first
n = AT 2 exp(/kB T )
(R1.1) sight be thought that, even if the anode were only
2
where A = 4kB
m/h3 is a universal constant which slightly positive, all emitted electrons would travel
has a value of 7.5 1024 m2 K2 s1 . Equation directly to it. In fact, unless the anode is typically
R1.1 is known as Richardsons law of thermionic at least a few tens of volts positive with respect to
the cathode, they do not. If its potential is lower
emission.
the density of electrons near the cathode (where
1 Classically, all of the electrons would acquire thermal
energies of this magnitude. That relatively few do so is a they are moving most slowly) becomes large enough
for their negative charge to repel further electrons
consequence of quantum restrictions on their energies.
94
EXPERIMENT R1.
back to the cathode. Equilibrium is reached when

the accumulated space-charge is sufficient to turn
back all but the few electrons necessary to replace
those lost from it to the anode. In such circumstances the anode current, I, is related to the anode
voltage, V , by Childs law:
THERMIONIC EMISSION FROM TUNGSTEN
1. Calibration
Before heating the cathode it is essential to obtain

a value for R20 by balancing the calibration circuit
shown in Figure R1.2. This shows a Wheatstone
bridge network (research how these work before
attempting the pre-lab test), with the nominal 1
3/2
resistor as a standard resistance and the two reI = fV
sistance boxes as variable and ratio arms. To enwhere the constant of proportionality, f , depends able precise measurements of R, a bench multimeon the geometry of the valve, but not on the number ter with sensitivity of 0.1 A is used.
of electrons emitted by the cathode. The current is
said to be space-charge limited.
C. Anode further positive: saturation
R1
Constant
current
supply
G
If the anode potential is sufficient to remove the
space charge, all electrons emitted from the cathode are able to reach the anode, and the current
depends mainly on the area, temperature and maR2
terial of the cathode. A slight increase of current
with increasing anode potential does occur, because
the potential barrier which the electrons have to
overcome to leave the vicinity of the cathode is reduced as the field increases. This phenomenon is Figure R1.2: The Wheatstone bridge arrangement
known as field emission see Bleaney & Bleaney. used for measuring the resistance of the cathode at
room temperature, R20 .
Data Collection
Overview
For various cathode temperatures, you will measure
the anode current as a function of anode voltage,
and determine the current Is emitted thermionically from the cathode. Note that Is is the saturation value of the anode current (region C). A
correction for field emission may be necessary assume it is proportional to the field at the surface of
the cathode.
By plotting ln(Is /T 2 ) versus 1/T , you will examine the validity of Richardsons law, and obtain an
estimate of the work function of tungsten.
The temperature of the tungsten cathode may be
obtained from measurements of its electrical resistance. For tungsten the relation:
R = R20 1 + + 2
Use your Wheatstone bridge and the low-voltage

power supply to measure R1 and R2 when the circuit is balanced (i.e. the potential difference, and
hence the current, across G are zero) for several
measuring currents. This measuring current is provided by the constant current supply and should be
between 100 mA and 1 mA.
Since these measuring currents should not heat
the cathode significantly above room temperature,
its resistance will remain constant at R20 . This
means that no anode current will flow (i.e. no electrons will escape the cathode due to the gain of
thermal energy), and the high voltage part of the
circuit which collects the anode current therefore
does not need to be connected.
Because R20 is typically a few tenths of an Ohm,
the voltage across it when 1 mA flows is of the order
of 100 V, so you will need to detect out-ofbalance voltages of 1 V if you are to measure
R20 within 1 %. The bench multimeter is easily
capable of detecting this.
(R1.2)
holds, where R20 is the resistance at room tem- 2. High Temperature

perature (actually 20 C); = T 293, where T Once you have obtained a value for R20 you can
is the cathode temperature in Kelvin; = 5.24 proceed to collect data at higher temperatures.
103 degree1 and = 7 107 degree2 .
Increase the current (switch to the 600 mA setClearly in order to use this relationship the value ting) through the cathode to increase its temperature. The electrical circuit of the apparatus is
of R20 must first be found.
95
shown in Figure R1.3. The cathode should still
be supplied from the 5 V terminals on the power
unit. Do not draw more than 0.6 A from the
supply, as this can damage the valve. Connect
the high voltage supply across the diode using the
shrouded cables provided. Remember to turn off
the power supply as you connect the circuit.
The electrons that are released thermionically
from the cathode and collected by the anode are
known as the anode current; this is displayed on
the 0 to 150 V variable supply. The anode current
should not normally be allowed to exceed 10 mA.
You will probably see the diode start to glow at
350mA, but you may not detect an anode current
until cathode currents of approximately 420mA are
applied.
When determining values of anode current, IA ,
and anode voltage, VA , you will need to take some
care in reading the meters. This is an experiment
where over-hasty work will produce particularly
poor results.
Appropriate cathode temperatures will be found
when heater currents are between 600 mA and
450 mA. You should collect the data described below for at least six cathode temperatures within
this range.
1. At each new temperature, first re-balance the
bridge circuit by varying R1 and R2 , to enable
you to calculate the new resistance.
2. Once the circuit is balanced, collect data of the
anode current versus anode voltage.
You are unlikely to need anode voltages above
50 V - at lower temperatures you will collect
all relevant data at anode voltages of less than
5 V. Hence careful experimentation is essential.
0.6 A
MAX
R1
5V
G
R2
mA
0-150 V
3. Check your results for reproducibility and plot

graphs of IA versus VA as you go along.
Figure R1.3: The Wheatstone bridge arrangement
used for measuring the resistance of the cathode at
Your measurements of R will all be in terms of high temperatures.
the nominal 1 resistor. Do you need to know the
precise value of its resistance?
The leads to the tungsten cathode will contribute
to the measured values of R. Can any correction
be made for this?
Be very careful to ensure good electrical connections in the resistance measuring circuits. Contact
and lead resistances can cause significant systematic errors.
Experiment R2
Heat Transfer
Aims:
1. To fabricate and characterise a temperature measuring device.
2. To plan, perform and evaluate an experiment to study rates of heat transfer.
Objectives:
1. To gain experience planning, evaluating and refining experimental methods;
2. To develop data analysis skills;
3. To develop understanding and make allowances for the practical factors to be taken into
consideration when planning experiments.
(c)
(b)
(d)
(h)
(g)
(a)
(f)
(e)
Figure R2.1: [Colour online] The experimental apparatus: (a) Thermocouple wire; (b) Multimeter; (c)
Fume extractor; (d) Soldering iron; (e) Stopwatch; (f) Safety glasses; (g) Solder; (h) Samples.
Safety Note:
This experiment uses very hot soldering irons, dry ice, water and ice near mains electrical equipment. Care must be taken at all times and beakers of water and dry ice must be placed in the
plastic tray provided. Do not start this experiment until you have read and signed the
risk assessment. Ask for help if you are unsure.
96
97
Introduction
For small changes in temperature, T , the

change in potential difference across a junction beIn this experiment you will construct and calibrate tween metals A and B, VAB , varies linearly with
two temperature-measuring devices known as ther- temperature, related by the Seebeck coefficient, :
mocouples, and you will then use them to investiVAB = T
(R2.1)
gate the thermal properties of objects - specifically
the rates at which they warm or cool.
This holds for small temperature changes only,
You should be able to calibrate your thermocou- since the Seebeck coefficient varies with temperaples, obtain all required data and complete a signif- ture.
icant amount of the analysis during the lab session
In general for thermocouples the relationship bebut you will need to plan your experimental proce- tween voltage, V , and temperature, T , is described
dure carefully.
by a polynomial function of the form:
Theory
Thermocouples
A thermocouple is a device which uses the current
which flows when two different metals are in contact in order to measure temperature.
In simplistic terms, metals are viewed as fixed
positive nuclei surrounded by a sea of free electrons. The free electrons are still attracted to the
positive nuclei, so energy must be applied to remove
the electrons from the system.
The amount of energy required to remove an electron entirely from the surface of a metal is known
as the metals work function.
If a metal is heated up, energy can be absorbed
by the electrons and they are more likely to escape
the attractive forces in the conductor. If one end of
a metal is heated up and the other remains cooler,
free electrons at the hot end will migrate to the
cooler end of the conductor, and there will be a
net negative charge at the cool end. This creates
a potential difference across the conductor, which
will eventually repel any additional incoming electrons, so an equilibrium will be reached. Generally
the potential differences produced are very small microvolts per degree Kelvin.
T = a0 + a1 V + a2 V 2 + a3 V 3 ... + an V n (R2.2)
where ai are polynomial coefficients, unique to
each thermocouple.
The metals you will be using have been chosen
because they have a relatively high Seebeck coefficient, and the potential difference induced varies
approximately linearly with temperature over the
range of your experiment - the Seebeck coefficient
remains approximately constant.
Figure R2.2 shows a diagram of the wires in the
thermocouple you will be constructing. Although
only a single junction between two dissimilar metals
(in this case A and B) is required to produce a voltage, measuring it requires connection to a multimeter, which will contain wires of a different metal, C
(usually copper). This would result in two more
junctions (A-C and B-C) and the introduction of
two extra voltages.
Instead, metal A is used to connect to both
points of the multimeter, so the extra voltages produced on either side will be equal and will cancel
out. This requires a second A-B connection, so the
voltage measured at the multimeter, V , will be the
total of the voltages produced at all junctions:
V = V3 + V1 VREF V4 = V1 VREF
(R2.3)
Using Equation R2.1 the voltage measured on

This effect of a temperature difference across a
conductor leading to the formation of a potential the multimeter is therefore:
difference was discovered by Seebeck in 1821, and
V = T1 TREF
(R2.4)
is known as the Seebeck effect.
When two metals with different work functions are in contact with each other, the differing strength with which the electrons are held
means that electrons will tend to flow from the
metal with the lower work function to the metal
with the higher work function - a potential difference is again produced.
Provided the reference junction remains at a constant temperature for all measurements the voltage
measured by the multimeter should vary linearly
with probe temperature.
Furthermore, the use of an ice bath for the reference junction should result in a zero voltage reading on the multimeter if the probe junction is at
Changing the temperature of the two metals re- the same temperature - i.e., the thermocouples can
sults in a change of the potential difference across be specifically calibrated to produce a zero reading
the junction.
at 0o C.
98
EXPERIMENT R2.
HEAT TRANSFER
V3
Multimeter
+
V
-
Metal C
Metal A
Metal C
Metal A
+
T1
Metal B
V1
-
TREF
V4
+
VREF
Figure R2.2: A sketch of the connections in a thermocouple circuit. The voltages produced across
junctions of dissimilar metals are indicated, and the voltage measured by the multimeter is V .
Heat Transfer
Energy is transferred due to a temperature difference through three distinct processes: conduction,
convection and radiation. An object placed in an
environment of a different temperature will lose or
gain heat in a combination of all three processes.
The conduction of heat depends on the objects
surface area, the temperature difference and a property of the material known as its thermal conductivity.
The convection of heat depends on the flow properties of the surrounding fluid.
The radiation of heat is proportional to the surface area of an object and its absolute temperature
to the fourth power - this is known as the StefanBoltzmann law.
The combination of these three processes is complex, however in all cases of heat transfer Newtons law of cooling states that the rate of cooling
of a body is directly proportional to the temperature difference between the body and its surroundings.
From this we can see that the change in temperature, dT , of an object during a short time interval dt is proportional to T , the difference between
the objects temperature and the environments temperature. The change in temperature is of course
equivalent to the change in temperature difference.
We will call the initial temperature difference (at

t = 0) T0 . Integrating, we find:
Z
T0
T0
dT
= k
T
t0
dt
(R2.7)
or
ln
T0
= kt0
T0
(R2.8)
where T 0 is the temperature difference between

the object and the environment at time t0 .
Dropping the primes and taking the exponential
of each side:
T = T 0 ekt
(R2.9)
Method
Fabricate two thermocouples from chromel and

alumel wires, as shown in Figure R2.3. You should
need no more than 30 cm for each length of wire,
and will need to strip 10-15 mm of shielding from
each end.
The chromel and alumel wires should be soldered
together to a point at the two junctions. Note that
the metal of the contacts within the multimeter is
unknown but you can assume it will be the same
dT = kT dt
(R2.5)
for each junction - therefore it can be ignored.
Here k is simply a constant of proportionality
which we will call the thermal constant. The negative sign is because an increase in time will lead Calibration
to a temperature change which will decrease the
Connect your thermocouples in turn to a multimetemperature difference. Rearranging we obtain:
ter using crocodile clips and check they both produce a voltage when the probe and reference juncdT
= kdt
(R2.6) tions are at different temperatures.
T
99
Probe
junction
Chromel
Extension
If you feel you may complete the analysis during

the lab, place an ice cube into some dry ice to cool.
Once you have completed the analysis for the
Chromel
main experiment, remove the ice cube from the dry
Multimeter
Test
ice and investigate its rate of temperature change
object
as it warms, then discuss your results.
Reference junction
If time is limited the ice cube can be removed
(Ice bath)
before it reaches the temperature of the dry ice,
but you should allow it to reach at least -20o C. If
Figure R2.3: The circuit used to measure the tem- the ice cube is removed early, discuss the effects
perature of an object at the probe junction.
this may have on your results.
V
Alumel
Ice cubes, dry ice, a heat stage and a thermometer are available - investigate the voltages produced
across one of your thermocouples over a wide range
of temperatures. Test your second thermocouple to
evaluate if a second calibration curve is needed.
Experiment
You have been given two identical objects made
from the same material, both with a hole into which
you should be able to fit your thermocouple probe
junction. You may need to use a small amount of
thermally conducting paste to ensure a good thermal connection between the junction and the object. Attach the thermocouples securely, then heat
one object in boiling water and cool one in dry ice
until the temperatures are stable.
Now investigate the rate at which the objects
return to room temperature. You will need to plan
your readings carefully.
Analysis
Your calibration data should be used to produce
an equation which will enable you to convert the
voltage across your thermocouple to the temperature of the probe junction. Is a linear fit to your
calibration data appropriate?
Next, plot a simple graph of thermocouple voltage against time for each object.
Convert the voltages to a temperature and plot
appropriate straight-line graphs to determine if the
temperature change varies according to the relationship described in Equation R2.9.
If this relationship is valid, determine the thermal constant in each case.
Note: you are encouraged to plot graphs in Origin, however you should always note the equations
used to calculate your column values and their errors in your lab book.

Physics Laboratory Handbook

Uploaded by

Copyright:

Available Formats

Physics Laboratory Handbook

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Physics Laboratory Handbook

Uploaded by

Copyright:

Available Formats

What is the purpose of the laboratory timetable and experiments?

What is the purpose of the laboratory timetable and experiments?

What are the requirements for students attending the laboratory sessions?

What are the requirements for students attending the laboratory sessions?

Level 4 Handbook

Practical Physics Laboratory

Level 4 Practical Laboratory Timetable: 2014 2015

A Guide to the Level 4 Teaching Laboratory

Code of Practice for the Teaching Laboratories

1 Introduction to the Level 4 Practical Physics Laboratory

2 A Guide to Good Laboratory Practice

3 Error Analysis: Important Concepts

4 How to Write Formal Reports

Measurement of the Speed of Light

K The Black Body Radiation Law

M Measurement of the Charge of the Electron

Q1 Photoemission of Electrons from Metallic Surfaces

Q2 Charge:Mass Ratio of the Electron

R1 Thermionic Emission From Tungsten

A Guide to the Level 4 Teaching

Code of Practice for the Teaching

Do not bring food or drinks into the lab. They

Good Lab. Practice & Standard

Report any breakages or spills to a technician

First, all undergraduate students must attend a

Contact details for first aiders are displayed near

For electrical equipment, isolate the power at

Bags and coats should be stored in the white

where possible block the beams so that they

Specific Hazards and Control

Liquid Nitrogen & Dry Ice - risk of burns and

The following items are all used frequently within

Numerous extremely hot materials - sources

Lasers - risk of eye damage from various sources,

All teaching lab. lasers should be Class II so

When leaving the labs

Version 2, created 30/07/13.

Introduction to the Level 4 Practical

features within the experimental apparatus. To do

The laboratory timetable is shown at the front of

The Laboratory Coordinator is Dr. Jaap Velthuis

The Physics Teaching Laboratories are governed by 1.3.2 Staff Demonstrators

CHAPTER 1. INTRODUCTION TO THE LEVEL 4 PRACTICAL PHYSICS LABORATORY

The staff demonstrators are happy to help with

Aims of the Level 4 Lab

A primary aim of your work in the Level 4 Lab is to

1.5. LABORATORY REQUIREMENTS

Each quiz may be attempted twice - a second

CHAPTER 1. INTRODUCTION TO THE LEVEL 4 PRACTICAL PHYSICS LABORATORY

Advised to repeat until 70% attained

Must pass - Edit until 6/20 attained

Must pass - extra sessions until completed

The final mark recorded for your Laboratory unit

attending labs or achieving deadlines difficult. In

Submitting Laboratory Work

Experimental lab books

Submitting Formal Reports

All formal reports must be submitted electronically

All written work will be checked electronically and

Attendance at all laboratory sessions is mandatory,