Faculty Handout for Fall 2006
The purpose of the Graduate Laboratory is to give students an in-depth experience in experimental physics where they can see how laboratory measurements are used to test ideas in physics. Students should learn how to make meaningful comparisons between their data and a theoretical model. They will have an opportunity to use and understand some contemporary measurement techniques. Each student will be required to complete two experiments from two different areas of physics and to produce a report in the form of a journal publication manuscript for each. These two experiments are the main focus of the course. However, the other aspects of the course will contribute significantly to the grade.
The following grade breakdown is more structured than has been traditional. The intent is to make expectations more explicit so that time and effort may be better managed. The percentage distribution is as follows:
All students will do two experiments. Some experiments are very challenging while others are more straightforward. Typically, the first experiment is the most difficult because you will not be familiar with Grad Lab and how it works, where things are, and what, exactly, you should be doing. Going through the complete cycle of the first experiment will prepare you so that you can be more ambitious on your second experiment.
There are eight parts to each experiment:
with the instructor
This is a 20 minute exploratory discussion of the
experiment you have chosen. Its purpose is to identify an appropriate measurement goal and
it will be a chance for you to ask questions and advice on how to get started. Reading the
papers referenced in the manual will help familiarize you with your options. The better
prepared you are the more you will get out of this discussion, but we understand that it
can be a challenge figuring out what's going on at first. This discussion is for your
benefit. It is mandatory, but not graded.
Supported by your lab notebook, you will present a short
discussion of the experiment you plan to do; what the physics behind the experiment is,
how and with what equipment you will perform the experiment, and what the statistical
design of the experiment is. You should be able to predict (though you might not be right)
the accuracy and precision of your expected results. You should already have begun to
familiarize yourself with the setup of your experiment before the proposal so that we can
discuss the experiment in as much detail as possible. But, do not use equipment until you
are confident that you know how to use it correctly. Starting an experiment can be the
most difficult part of this course so don't be afraid to ask questions and for help! The
proposals will be graded.
Taking data is generally the easiest and fastest part of Grad Lab experiments. Most of your time will be spent in familiarizing yourself with your experiment, setting up your apparatus, and testing (and calibrating!) equipment. To receive credit for doing the experiment, you must demonstrate and explain the working experiment to the instructor. It is your responsibility to arrange with the instructor a time to do this.
You are required to keep a bound page-numbered research notebook in which all your work on your experiments is described in ink! The notebook should be a "Computation Book" with large 9 1/2" x 11 5/8" numbered quadrille pages or equivalent. DO NOT BUY THE SIMILAR LOOKING LAB NOTEBOOK WITH WHITE AND YELLOW PAIRED PAGES WHICH TEAR OUT EASILY. IT IS FOR LAB NOTE SHARING AND IS NOT SUITABLE FOR THIS COURSE. Keep this notebook as a research journal, in chronological order with each entry dated. Include a summary of your reading and study on experimental background; experiment design; preliminary calculations and error studies (What will be the most significant errors?); description and characteristics of the instruments used and their calibration; analysis of the procedure and first data taking to check feasibility; detailed recording of the final data with enough detail of gains, equipment, etc., to allow you to return to where you left off in a minimum of time.
This notebook is the detailed record of your experiment. Learning to keep a research journal is an important skill as a scientist and essential to doing a good experiment. Do not neglect it. It will be graded.
Because students may be sharing equipment you need to be both considerate and assertive. If you leave equipment running that may need to be shared, leave a dated note with your name, email address, and phone number.
The purpose of doing an experiment is to obtain the
best results possible, subject to the constraints of time and equipment. You should
maximize your effort toward these goals. In particular, it makes little sense to devote a
great deal of time and effort to measure a value to 1 part in 104, if you know the calibration is no better that
There will be weekly progress reports as well as the
mid-experiment progress reports scheduled on the calendar. They will count as part of your
in-lab work grade. These are meant to help you. Students sometimes try to do too much,
focus on the wrong aspects of experiments, or stop making efficient progress, and don't
realize that we can help if we know what their situation is. The quality of these reports
is not graded. You get full credit for simply having a meeting with the instructor. The
nature of the report depends on the experiment, but will, at least, consist of a review of
You will demonstrate a working experiment to the professor and/or the TA at a mutually agreed upon time. The discussion should be quantitative and detailed. Demonstrations will consist of:
These demos can take anywhere from 10 min. to an hour depending on the experiment and your preparation. For experiments requiring special preparation, such as liquid He, arrangements must be made in advance. It is best to show real signals from the experiment.
Near the end of each experiment, you will give a qualitative
explanation of your experiment to your classmates. The entire group will do this together
by going from station to station. We will select one of the lab days for this event. Note:
Class conflicts will require that you miss your other, conflicting class, dependent on
the time needed. You should prepare to talk for between 10 and 15 minutes. If it is
possible to show live signals from your experiment, please do so. But, the purpose of this
exercise is to give you the chance to explain what you are doing and for others to benefit
from your new-found knowledge. The tour need not be as comprehensive as the demo
for the instructor.
The "rough draft" is a chance for us to give you
feedback on what your final paper should be like before you spend inordinate amounts of
time writing it. You should have your Figures, Captions, and Tables for the paper ready
for this draft presentation. An outline of the paper would also be helpful. You don't need
to have all of the writing done, but useful comments could be made about the procedure and
discussion sections of the paper, if they were ready. This should help you spend your time
more effectively. The rough draft is not graded, but it is essential if your final paper
is going to meet my expectations. The defense is just the process of supporting your draft
with you notebook as support for your calculations, uncertainties in results, and figures.
The final paper should include a summary of the relevant theory, description of how the experiment was performed, data and analysis, in-depth discussion of errors, results, and conclusions. The length of the theory section should be limited to a few pages at most. The paper should be typed in the manuscript style of an article for a journal of The American Institute of Physics. It must contain appropriately footnoted references. (See the Graduate Laboratory Handbook for further details on format). The tone of the paper should be that of an experimental physicist performing a "real" experiment. The emphasis is on the quality and validity of the work, not on getting the right result. Disagreements with simple theory or other results should be discussed in light of experimental technique and uncertainty, as well as potential higher order theories. Often the small, but statistically significant, bump is the most interesting part of the experiment.
Presenting your results with a clear description of a valid error analysis and major errors is essential. It is also essential that your notebook contain the original data and analyses supporting your: results, uncertainties, tables, and figures of the draft and written papers. Work done in spread sheets, plotting, and fitting programs needs to be documented with file names, descriptions of calculations, and conclusions. Plots or at least reference to hard copies of plots in folders are also useful.
The facilities, regulations, and organization of the
Graduate Laboratory will be described by Mr. Monroe and the instructor. The data analysis
book, Data Reduction and Error Analysis for the Physical Sciences (Second Edition)
by P. R. Bevington and D. K. Robinson is a required book for the course. (A useful
reference for a more elementary treatment of error analysis is An Introduction to Error
Analysis by J. R. Taylor.) The Bevington text includes the Turbo Pascal algorithms
used in the Graduate Laboratory data analysis programs LINREG and ANALYSIS.
The second required book is the American Institute for Physics Style Manual. The Style
Manual is now available on the web at http://www.aip.org/pubservs/style.html .
This manual give a full description of the format details of preparing manuscripts
for publication and will be the style that you will use for your reports in the
Data analysis and plotting must be done on the PC's in the
laboratory. Approval must be obtained to use other computers for these tasks because
outside programs often do not do the required statistical analysis. The Graduate
Laboratory has a selection of analysis and plotting programs available. We will be glad to
give you any help you need learning to use these programs. See the appendix of the Grad
Lab Manual about our software.
Regular communication is essential in Grad Lab. Besides the required attendance on your scheduled lab day, E-mail is the next easiest way to send messages. You are expected to check your e-mail regularly, at least daily.
To facilitate communications please list your office,
office-phone, e-mail address, and home phone numbers on the information cards that will be
circulated to the class members. Let me know of corrections needed to this information on
the copy once it appears on the bulletin board. Note that the "Student Roster"
form on the web pages provides phone and email information for students who may be sharing
Listed below are the available experiments.
We may require you to defer your selected experiment to permit orderly scheduling; there is only one copy of each experiment. In general, it is preferred for only one student to perform any experiment from each lab section. Exceptions for which two students may be permitted to sign-up are denoted with a "(2)". The experiments must be done independently, even if sharing the equipment. Scheduling must be arranged so that each student working on a two-student, shared experiment has time on the apparatus during the scheduled hours of the class for the purpose of discussions with the faculty and TA.
Numbering of the experiments is in accord with the Graduate Laboratory Handbook.
Diversification is required you may not do all your experiments in low-temperature physics, for example.
For queries regarding: