Mechanical Engineering (ME) 4201

Mechanical Engineering Design Laboratory

Lab Manual

Compiled/Prepared by:
Department of Mechanical and Industrial
Engineering
Louisiana State University

Fall 2024
 TABLE OF CONTENTS

PREFACE ....................................................................................................................... 3

REPORT REQUIREMENTS ........................................................................................... 4

Introduction ................................................................................................................. 4

Format ........................................................................................................................ 5

General Requirements ................................................................................................ 9

Report Contents ........................................................................................................ 10

Writing Tips and Recommendations ......................................................................... 14

(A). STATIC AND DYNAMIC BALANCING EXPERIMENT .......................................... 17

(B). CAM/FOLLOWER EXPERIMENT ......................................................................... 21

(C). SECOND-ORDER SYSTEMS EXPERIMENT....................................................... 25

(D). STRAIN GAUGE EXPERIMENT ........................................................................... 33

(E). JOURNAL BEARING LUBRICATION EXPERIMENT ........................................... 39

(F). VIBRATION ANALYSIS OF ROTATING EQUIPMENT EXPERIMENT ................. 47

(G). RESONANCE EXPERIMENT ............................................................................... 59

(H). FIBER-BRAGG GRATING EXPERIMENT ............................................................ 67

2
 PREFACE

This manual reflects the work of multiple contributors over many years. It is
constantly being updated to reflect changes to the experiments, to correct errata, to
improve clarity, and generally to improve the experience of students taking this course. If
you find anything incorrect or misleading in the document, please let us know. If you use
any information or images from this manual, please remember to properly cite the authors
(the MIE department).

After each experiment, your group will write and submit a formal group lab report.
The following section covers report requirements specific to this course. Be sure to read
these carefully.

For each experiment, you will find a brief discussion of some of the topics and
theory involved, followed by step-by-step instructions and any specifically required
analysis. The discussion and theory sections are not exhaustive; this is intentional. One
of the requirements of this course is to demonstrate an ability to formulate and present
theory and relevant examples. You will achieve this through a combination of the
following:
• Drawing free-body diagrams and using them to explain the meaning of formulas;
• Providing any missing formulas needed for the analysis;
• Providing definitions of key terms; and
• Researching and presenting real-world examples that highlight the relevance of
the topics you are discussing.
Any necessary theoretical principles not covered in the lab manual have been covered in
your previous core courses and/or can be easily found using online resources.

3
 REPORT REQUIREMENTS

Introduction
The importance of good report writing and data presentation cannot be
overemphasized. No matter how good an experiment, or how brilliant a discovery; it is
worthless unless the information is communicated to other people. As you complete each
laboratory report, you will be gaining experience in writing technical reports and will be
developing skills you will find invaluable throughout your career. There are a great many
books available on report writing in addition to the resources provided here.

You will be providing formal written lab reports for the experiments performed in
this course. As such, you are expected to know and follow the general guidelines covered
in the Moodle course ME Undergraduate Lab General Guidelines. The following are
specific requirements for the reports in this course. Please read them!

Purpose
The lab reports for this course are the primary means of assessing student
performance. As you prepare your reports, keep in mind that it should achieve the
following goals:
1. Present your experiment and results with enough detail to be clear to the reader;
2. Demonstrate a deep understanding of the basic engineering principles involved,
and your ability to communicate these principles clearly and accurately;
3. Demonstrate accurate and appropriate engineering calculations;
4. Demonstrate your ability to apply engineering knowledge, calculations, and
statistical techniques to real-world situations;
5. Demonstrate your ability to draw meaningful conclusions from real-world data;
6. Demonstrate your critical thinking skills, including identifying likely sources of error
and making practical recommendations for improving experimental techniques;
7. Demonstrate effective technical communication skills – both written and visual;
8. Give you practice to continuously improve your communication skills to better
prepare you for other courses at LSU and for your future engineering career.

Audience
For the purposes of this course, you should write your reports for the following
audiences. A portion of your grade is based on effective communication.
1. The primary audience should be assumed to have some background in science
or engineering, but have no specific knowledge of the experiment in question.
Therefore, you need to explain the engineering principles being investigated, how
the experiment was performed, what the results were, what the results mean, and
why they are important. The report should also be detailed enough that an
engineer at another facility could select appropriate equipment, recreate the
experiment, and obtain similar results to validate your own.
2. The secondary audience in this course will be the grader. Therefore, you also
need to demonstrate proper completion of the lab requirements, sufficient
understanding of the theory and its application, and the ability to efficiently
communicate information.

4
Length
There are no set requirements for report length for this course. However, if the
main body of your report (i.e., Introduction to Conclusions) exceeds 10 to 12 pages
(double-spaced), consider that you may be giving too much detail, repeating information,
or need to reword or reorganize parts of your report to condense it. If your grader
considers your report to be excessively long or wordy, you may lose points because of it.

Format
Please obey the following guidelines when finalizing your reports.
• All formatting should be reasonable and consistent throughout the report.
• Fonts should be easy to read, with a 10- to 12-point font size. This includes fonts
in tables and figures, wherever possible. Section headings may use a larger font,
but should not be excessive.
• All text should be double-spaced, to allow room for grader marks and comments.
Exceptions may be made for tables or captions, or elsewhere if circumstances
dictate.
• Margins should be applied consistently throughout the report. Recommended
margins are 0.75” to 1.00” all around. No text, tables, or figures should extend
beyond the page margins.

Narrower margins may be used if necessary for oversized scanned images, such
as the cover page or scanned pages in the appendices. Use the narrower margins
only in these sections, and only if it cannot be avoided.
• Paragraphs may use block format (with a space between paragraphs) or indented
format (with no space between paragraphs). This is your choice, but you must be
consistent throughout the report.
• The Table of Contents, the Introduction section, and each appendix should each
start on a new page. For all other sections, start immediately after the previous
section. Do not jump to the next page except to avoid orphan lines.
• The cover page and title page should not be numbered. All other pages must be
numbered at the bottom of the page. Page numbering may follow one of two
formats:
o Roman numeral front matter. The Table of Contents, List of Figures, and
List of Tables should be numbered with roman numerals, starting with page
ii (i.e., the title page is implied to be page i). The main body and appendices,
starting with the Introduction, should be numbered with Arabic numerals,
starting with page 1. This format makes it easier to find the beginning of the
main body and to keep track of its page count.
o Continuous numbering. The Table of Contents begins on page 2, and
pages are consecutively numbered through the rest of the document. This
format is more common in short documents and requires less work to set
up.
• All scanned pages should be inserted at a 1:1 size ratio with the originals, and
should be undistorted and easily read. Figures 1 – 3 demonstrate unacceptable
and acceptable scanned images. Note that it can be difficult to obtain acceptable
results with a cellphone camera. Use a scanner if needed to obtain a better image.

5
Figure 1. Sample scanned page. The image is too small, making the text difficult to
read. This is not acceptable.

6
Figure 2. Sample scanned page. The image is distorted and unevenly lit, making it both
difficult to read and unprofessional looking. This is not acceptable.

7
Figure 3. Sample scanned page. The size ratio is 100 %, and the text is clear. This is
an example of an acceptable scanned image.
8
General Requirements
Writing Style
Third-person past tense (“The data were recorded…”) is generally accepted as the
preferred grammatical style for technical and scientific reports, and is required for this
course. Use third-person present tense for statements that continue to be true into the
present (e.g., “Accelerometers are used in the automotive industry…,” “Figure 3 shows
the recorded data…”). Do not include commands (the imperative mood, e.g., “Turn the
motor on.”) in your lab reports, as these are not third-person.
Follow all rules for formal writing. Avoid contractions (“can’t,” “don’t,” etc.) and
slang terms. Technical jargon is appropriate, but define any terms your audience may
not know.

Figures, Tables, and Equations

All figures must have a numbered caption, placed below the figure. All tables must
have a numbered caption, placed above the table. All equations should be numbered,
preferably with right-justified numbers. This is necessary so you can refer back to
previous equations later in the report.

Probability and Statistics

Probability and statistics play an important role in experimental work – be it
academia or industry, research or production. Data presented without consideration of
statistical/uncertainty analysis can be misleading because measurement errors can skew
the results. Therefore, you are expected to use the following analysis techniques in your
reports.
1. If a measured value is visibly fluctuating, record the level of fluctuation on your raw
data sheet – either as a mean and ± value or as a min. – max. range. This counts
as a source of uncertainty in the measured value. This can mimic a 95 % precision
interval – for example, if a value is between 1890 RPM and 1930 RPM at least
95 % of the time, but occasionally jumps to 1970 RPM, you can record it as
1910 RPM ± 20 RPM. Don’t forget to report uncertainties with your results.
2. If more than one measurement is taken for a given value, include the mean and
confidence interval (or precision interval, if appropriate) wherever the data is
reported. In tables, typically devote one column each to mean and uncertainty. In
graphs, use error bars to indicate uncertainty.
3. If you use a curve-fit or regression, include the equation of the curve and the
correlation factor (R²). In some cases, you may need the two curves representing
the 95 % confidence interval (Scheffé band) to demonstrate the goodness of fit or
to make comparisons between two fits.
4. When comparing an experimental value to another value or looking at trends,
uncertainty analysis is needed for meaningful interpretation. You are expected to
recognize when and where this is needed, and to perform the analysis. Order-of-
magnitude approximations may be used when sufficient. Full uncertainty analysis
may be needed in some instances.

Outside Sources
A few sections of your report will require additional research or supporting sources.
The Introduction and Background can both benefit from having reliable sources to back
9
your assertions. You may need to cite sources for the more obscure equations in your
Theory section. You might cite the lab manual when writing the procedure, or use images
you find online. In the Conclusions, you will show how some aspect of the experiment
applies to a real-world scenario. Good sources add credibility to your claims.
One of the course objectives is to perform research to obtain new information and
incorporate it into your report. Therefore, you are expected to spend at least 10 – 20
minutes searching for information to help satisfy the report requirements. A minimum of
one outside source (other than the lab manual) is required for each report. A portion of
your grade will be based on the appropriateness of your sources and on how well you
incorporate the information into your report. Remember that Wikipedia should NOT be
considered a reliable source.
Lab reports and lab manuals from past semesters are not authorized materials.
Any student suspected of copying from or using old lab reports will be reported to LSU
Student Advocacy and Accountability, who will investigate and determine the outcome.

Report Contents
Your reports for this course should contain the following sections, in order. You
do have some freedom in splitting sections. For example, you may write the background
and theory as two separate sections, or as one.
Remember that everything in your report should serve to answer at least one of
the following questions: What was done? Why was it done? How was it done? What
were the results? What does this mean? Why is it important?

Cover Page
A “Cover Page & Assessment Sheet” is provided for the formal reports. Each
group must complete the top portion of this form – experiment title, section and group
number, author names, last four digits of LSU ID (in case of ambiguity), and the agreed-
upon percentage contribution of each group member. Scan the completed form and
insert it at 100 % of the original size as the cover page in the report.

Title Page and Abstract

Immediately after the cover page is the title page (page 𝑖 or page 1, unlabeled).
Include identifying information such as the experiment title, the authors’ names, and the
date written or submitted.
At the bottom of the title page should be the abstract. This is a short (typically one
paragraph) summary of the entire report, and thus should be written last. The abstract
should include what you consider the most important experimental results and
conclusions. (N.B., your report is not hidden behind a paywall, so there is no need to
entice readers with vague results. Provide definite, quantified results in the abstract.)

Table of Contents
A table of contents is required. Note that Microsoft Word can automatically
generate one for you if you use heading styles for your section titles.
Since your report will include figures and/or tables, a List of Figures and a List of
Tables are also required. These should follow the same format as the table of contents.
Again, Word can automatically generate these if you use the built-in caption tools.

10
Introduction
This section should introduce the topic(s) involved in the report and briefly explain
their relevance, necessity, and/or applications in engineering. You are expected to do a
small amount of research for the introduction/background sections and find reliable
sources and real-world examples to highlight the importance of what you are going to
discuss. Be as specific and illustrative as possible. For example, “Gears are used in
almost every engineering industry,” is vague. “Spur gears are used for everything from
drive transmission in bicycles and aircraft engines to producing the high torque needed
to run sugar mills,” is more illustrative, and piques the reader’s interest. You may also
find examples that highlight the importance of the subject matter at hand, and make the
reader care about the subject matter. For example, “A case study from 2012 showed that
a training aircraft lost an engine and was forced to make an emergency landing because
stress concentrations at the root of a drive gear, coupled with contact fatigue, caused a
crack to form, which ultimately led to part of a tooth breaking off and causing total failure
of the gearbox.”
This section should also briefly introduce the experiment and your main goals (i.e.,
what you are trying to measure, determine, or prove). The details will be giver later in the
report.

Background and Theory

The background is largely a more in-depth continuation of the discussion from the
introduction. Provide any information the reader is expected to know before reading the
rest of the report.
The theory is where you present and explain the engineering theory that will be
used in your calculations. For this course, the following are required in this section:
• All equations used in the analysis (i.e., the calculations) must be presented here.
In addition, all variables used in the equations must be clearly defined, and units
given where necessary. Note that simple equations all engineers would know
(𝐹⃑ = 𝑚𝑎⃑, mean and standard deviation, etc.) may be omitted.
• You should explain how each of the equations was obtained. Simply repeating
what the lab manual says will get you minimal points. Most of the equations
you will use come directly from first engineering principles, and your explanation
should demonstrate your understanding of this. A full derivation is neither
necessary nor appropriate here, however. A brief explanation in words and/or
diagrams should suffice. For more complex equations (e.g., Sommerfeld
equation), you may cite a reliable source.
• At least one free-body diagram is required in the theory section. You should use
this diagram to help explain the meaning/derivation of some of the equations used.
This is also a good place to define any coordinate systems used. Free-body
diagrams may be hand-drawn, but should be reasonably neat with all text legible.

Equipment and Setup

All equipment used in the experiment should be listed here. Do not simply copy
the list out of the lab manual. Include any relevant details – part numbers, dimensions,
etc.. You should also specify the use or purpose of equipment if it is not obvious. For
example, “Fluke multimeter – used to measure nominal resistance of the strain gauges.”

11
 This section must also include a diagram or diagrams showing the experimental
setup. You may take photographs to use in place of diagrams, but plan them carefully
and make sure they clearly show what you want the reader to see.

Experimental Procedure
List what was done in the lab, step by step. Include just enough detail that an
engineer at another institution could recreate the experiment if necessary. This also
allows the reader to identify any potential problems or sources of error in your
experimental technique. This section is typically most effective if written in the form of a
chronological list, as shown below. Short paragraphs may be allowed when appropriate.

1. A beaker was filled with 500 ml of water, and placed on a hot plate.
2. A thermocouple was placed in the beaker, and the temperature of the water
sampled once every 8 seconds.
3. The recording ...

DO NOT copy the procedure directly from the lab manual. Write primarily from
your own logbook and combine, split, or summarize steps to make the procedure shorter
or easier to read. Using tables for computer settings or other parameters may save time
and space. This is also the appropriate place to discuss any deviations from the intended
procedure. For example, if you originally intended to monitor the temperature of two
beakers of water, you might note that only one thermometer was available, and that this
was not possible. Writing down anything other than what was actually done in the lab is
tantamount to data falsification, and will result in lost points at a minimum if it is
discovered.

Results and Discussion

Present all results here – including numbers, tables, and graphs. This includes,
but is not limited to, the results of all analysis specified in the lab manual, including
uncertainty analysis. Introduce and briefly explain or comment on all tables and graphs
in the text; do not simply place a figure here and expect the reader to read it.
Do not show the analysis (i.e., the calculations) in this section. You should,
however, let the reader how the data were calculated (for example, “The manometer data
were converted to mean differential pressures using Equations 3 and 4, and are shown
in Figure 2.”) Refer back to equations in the Theory section as needed.
Brief observations and discussion of the results should be given here.
Observations should be quantitative whenever possible (e.g., “the pressure was 3 %
higher than the theoretical value,” “the two values are within 300 kPa of each other”).
Avoid vague observations (“the values are all close to each other,” or, “the two curves
follow the same general trend”) and unsubstantiated claims (“the pump behaved as
expected,” when no expectations have been given).
Any additional information relevant to interpreting the results may be given here,
as well. For example, you might mention that a particular data point exhibited significantly
more noise than the others did. You might state that the setup made it difficult to trace a
curve accurately, to let the reader know that this is a potential source of error.

12
Conclusions and Recommendations
Very often in life, your audience will skim the abstract and then jump to the
conclusions, only reading more of your painstakingly well-crafted report as needed.
Therefore, you may briefly summarize the results here, though it is not necessary. If you
do, keep it as concise as possible.
The primary purpose of this section is to interpret the results presented earlier and
distill them into useful information that was gained from the experiment. This is where
you showcase your understanding, your critical thinking skills, and your ability to apply
concepts to real engineering problems. EVERY conclusion you draw should be directly
supported by the evidence given. A few examples of the types of conclusions you can
draw include:
• Determining whether a theoretical expectation was met, using a clearly-defined
engineering criterion (such as from uncertainty analysis);
• Identifying the most likely sources of error1 (evaluating equipment, procedure, and
results) and determining how much these impacted the results;
• Quantifying functional relationships (“The mass moment of inertia appears to follow
a second-order relationship with respect to the flywheel diameter”).
• Extrapolating the techniques used to other engineering applications (“This indirect
method of determining mass could be used, for example, to predict the dry weight
of underwater wreckage, which is difficult to weigh directly because…”).

You are expected to show how your results, your technique, or anything else you
learned during the experiment could be applied to a specific real-world situation (such as
in the example above). This shows not only that you understand the material, but also
that you are beginning to think like an engineer.
Finally, you are expected to make a critical assessment of the experiment
performed and make realistic recommendations on how the equipment, procedure, or the
experiment in general could be improved. Again, all recommendations must be justified
by your observations or results. For example, if it was hard to draw conclusions because
of large uncertainty values, you might identify the largest source of uncertainty and
recommend a new piece of equipment or a different measurement technique to address
it. You may have an idea for making part of the experiment run more efficiently, or you
may point out a malfunctioning piece of equipment that needs to be repaired or replaced.
Insightful student comments are taken into consideration, and can actually help
improve the experiments moving forward. Comments such as, “Electronic pressure
sensors would be more accurate than the manometer,” or, “This experiment taught the
students about gyroscope couples,” without specific justification are less useful, and less
likely to earn you full points.

References
A formatted list of references for all sources of non-original material is required.
Any national or international standardized format is acceptable, though IEEE is a
convenient format for engineering documents. All sources should be cited in the report

1 Avoid ambiguous terms such as “human error.” If you suspect human error affected your
results, be specific on what exactly you believe you did wrong.
13
where the materials are used. Sources listed here but not cited in the report will be treated
as plagiarism and incur a penalty.

Appendices
Two appendices are required for each report:
• Logbook – the raw data sheet(s) and the procedure log recorded on the day of
the experiment, both with the TA’s signature or other documented form of approval;
• Sample calculations – at least one sample calculation per equation used in the
analysis, following the ME Undergraduate Lab General Guidelines.
These materials should be scanned and inserted at a near 1:1 ratio with the originals.
You may include additional appendices for information you feel is relevant but not
vital to the main body of the report, such as detailed derivations of equations or
supplemental plots not directly related to the discussion. The report should be complete
and coherent without the appendices, however. Include all important results and plots in
the main body.

Writing Tips and Recommendations

Preparation
• Before coming to the lab, be sure to read through the lab manual and complete the
training quiz to see what concepts or equations you may need to use. Brush up
on these before you arrive to help save time during the lab.
• In addition to generating blank data sheets, by knowing ahead of time what you
will be doing, you can start to write the Introduction and the Background & Theory
sections before you even start the lab. Note that you may have to make some
changes afterwards, however.

Documentation
• You need to record all raw data and your experimental procedure in your logbook
as you perform the experiment. For a group of three students, you might have one
or two students actively run the experiment, and the other(s) record the data and
procedure. Communicate clearly to ensure nothing is left out.
• Note that the TA must look over and approve both parts of the logbook before you
leave. This may be done with a signature or by photographing the pages.

Planning
• Before you leave for the day, sit for several minutes as a group and plan the report,
as well as everybody’s tasks.
o Make an outline of the report.
o Decide what material needs to go in each section (calculations, topics to
address in the Background and Theory, etc.).
o Decide what photographs or diagrams would be helpful – plan and take any
photographs before leaving the lab.
• It may be helpful to assign writing roles:
o Subject matter experts write about what they know best, perform
calculations, and/or generate graphs and numbers.

14
 o The lead writer assigns tasks, collects ideas, and writes a larger portion of
the report. They may also be responsible for keeping other team members
on schedule.
o The editor combines the material written by others, ensures the writing style
and information is consistent and coherent, and works on formatting issues.
• However you organize the writing process, ensure that the work is distributed
evenly, that all team members meet their obligations, and that everyone’s input is
present in the final document.

Writing
• Once you have the information you need, organize your ideas in your head or on
paper, and start writing.
• Don’t try to use big words or flowery language. Keep it professional but concise.
Write as if you were explaining this to someone in person; this may help the ideas
flow better. AFTER you finish your initial draft, then go back to correct the
grammar, reword informal writing, add correct engineering terminology, and such.
• Communicate early and often within your team to make sure you are writing a
coherent report. If possible, you may even write in the same room so you can
quickly ask questions or compare ideas.

Compiling the Report

• Have a working draft of the report from Introduction to Results and Discussion AT
LEAST 48 hours before the deadline. Combine the sections, and have the entire
group read it for accuracy and consistency. Discuss any changes that need to be
made at this point.

Once you have the rest of the report written, only then should you write the
Conclusions section. After that, read through the report once more and write the
abstract.
• Once the report is finalized (hopefully with at least a few hours before the deadline),
send a copy to each group member, so they can look it over and point out any last
minute corrections that need to be made.

Remember, each member is equally responsible for the submitted report as a

whole.

Getting Back the Graded Report

• After you get the report back from the grader, don’t just look at the grade and move
on. Read all the comments made, and identify whether or not you are making the
same mistakes in your next report. Be sure to ask questions about any comments
you don’t understand, so you can get a good idea of what the grader is looking for.
• Remember, a major purpose of the written reports is to help you sharpen the
communication skills you will use throughout your academic and professional
careers. Use the feedback wisely.

15
[This page intentionally left blank.]

16
 (A). STATIC AND DYNAMIC BALANCING EXPERIMENT

Introduction
Rotating machinery is a part of everyday life in industrialized society, from
automotive crankshafts and wheel axles, to pumps and turbines, to household fan blades.
When these rotating components are not properly balanced, dynamic forces are
introduced that can lead to anything from squeaking bearings or unwanted vibrations to
catastrophic failure. Being able to diagnose and correct imbalances is an important part
of maintenance for these components.

Objective
In this experiment, students will…
• Demonstrate the principles of static imbalance and dynamic imbalance
(sometimes known as couple imbalance), and able to explain the difference
between the two;
• Identify characteristic frequencies of a rotating system by analyzing the frequency
spectrum (discrete Fourier transform) of accelerometer signals;
• Investigate the effect of imbalances in rotating machinery on structural support
loads.

Necessary Equipment
• Machinery fault simulator
o Rotating shaft supported by two bearings
o Variable speed motor with electronic speed controller (ESC)
o Two rotating disks, each with two sets of ¼” threaded holes spread
circumferentially around the rotating disk – radius alternating between 2.25”
(5.72 cm) and 2.75” (6.99 cm)
• Four piezoelectric ICP® analog accelerometers (model 603C01)
• PC with data acquisition board
• Various masses
o 2× 10.0 g bolts (black)
o 8× 9.4 g bolts (silver)
o 12× 1.0 g washers
o 1× 0.28 g wire spring

Procedure

Part 1
1. With the computer powered up, launch LabVIEW and open the virtual instrument
(VI) file FaultSimulator.vi (found in the Documents folder). Run the VI and leave it
running.

2. Ensure no additional masses are attached to either disk. Close and latch the
plastic hood.

3. Start the motor and set the speed to 15 Hz using the electronic speed controller.

17
4. In the VI, measure the spectral analysis of the relative accelerometer data at the
rotational speed (15 Hz). In the “Cursor Values” box, set the Frequency control
value to 15. Then record the amplitudes of Accelerometer 2 with respect to 1, and
of Accelerometer 4 with respect to 3.

5. Throughout the experiment, pay attention to the Spectral Analysis graph, and
record any other peaks that seem noteworthy. Comment on these in your report.

6. Increase the motor speed to 30 Hz, and measure both relative accelerometer
amplitudes at a frequency of 30 Hz. Do this again for a speed of 45 Hz.

7. Stop the motor. Open the plastic hood, and screw a 10 g bolt into Disk 1 (the disk
between the two bearings). Ask your T.A. whether to use the inner (2.25”) radius
or the outer (2.75”) radius. Close and latch the plastic hood.

8. Repeat Steps 3 – 6 with the unbalanced disk, recording the amplitudes each time.

9. Stop the motor and open the hood. Remove the 10 g bolt from Disk 1 and screw
it into Disk 2 (the overhanging disk), using the same radius as in Step 7.

10. Repeat Steps 3 – 6 again, recording the amplitudes.

Part 2
11. Stop the motor and open the hood. Remove the 10 g bolt. You will now screw two
masses into one of the disks. Your T.A. will tell you the masses, the locations, and
which disk to use.

12. Close and latch the plastic hood. Start the motor and set the speed to 25 Hz.

13. Set the cursors to 25 Hz, and record the relative accelerometer amplitudes for 2/1
and 4/3 for the unbalanced case.

14. You will now attempt to re-balance the assembly by adding one additional mass to
the same disk. Calculate the required balancing mass and location using both the
graphical and analytical methods described on page 20.

15. Once you have calculated the required balancing mass, add it (or as close as you
can come) to the original disk.

16. Close and latch the hood. Start the motor, still at 25 Hz, and record the relative
accelerometer amplitudes.

17. Stop the motor and open the hood. Keep the two masses from Step 11 on the
original disk, but remove the balancing mass and move it onto the other disk.

18. Close and latch the hood. Start the motor and record the accelerometer data.

18
 Part 3
19. Stop the motor. Open the hood and remove all masses from both disks. Place a
single 10 g bolt on the outer radius of Disk 1. Close and latch the plastic hood.

20. You will now investigate the behavior of the rotating system over a range of
frequencies. Start the motor and set the speed at 30 Hz. Set the cursors to 30 Hz
and record the relative accelerometer amplitudes (2/1 and 4/3) from the Spectral
Analysis.

21. Increase the motor speed to 31 Hz. Set the cursors to 31 Hz and record the
amplitudes. Repeat for 32 Hz, 33 Hz, etc. up to 45 Hz, in increments of 1 Hz.

22. Stop the motor. Open the hood, and remove all added masses from the system.

23. Measure the following dimensions:

o Distance between the center of the left bearing (where accelerometers 1
and 2 are) and the center of Disk 1.
o Distance between the center of the left bearing and the center of the right
bearing.
o Distance between the center of the right bearing (where accelerometers 3
and 4 are) and the center of Disk 2.

24. Once you have verified that you have collected all necessary data, close LabVIEW.
If yours is the last section of the day, shut down the PC.

Report discussion
Please include the following in your report, in addition to any other relevant
information.

• For Part 1, discuss the effect of rotational speed and whether or not the results
match your predictions.

• For each unbalanced case in Part 1 (Disk 1 unbalanced at 15 Hz, 30 Hz, and 45
Hz; Disk 2 unbalanced at 15 Hz, 30 Hz, and 45 Hz), calculate the following:
o The centrifugal force exerted on the shaft by the rotating bolt,
o The additional reaction forces at the left bearing (𝑅2/1 ) and at the right
bearing (𝑅4/3 ) that would be needed to counteract this centrifugal force.

• For Part 2, discuss the differences observed between placing the balancing mass
on the same disk versus the opposite disk as the initial unbalance. Can you explain
these differences? What do these situations correspond to in terms of static and
dynamic balance? Be sure to define both terms in your report’s Theory section.

• For Part 3, what did you observe as the rotational speed changed? Were there
any unexpected results? Can you explain any phenomena that were observed?

19
Calculating the balancing force
1. Find the centrifugal force exerted by each mass on the rotating shaft,
𝐹⃑𝑐,1 = 𝑚1 𝑟⃑1 𝜔2 , 𝐹⃑𝑐,2 = 𝑚2 𝑟⃑2 𝜔2 , … (1)

2. Calculate the net centrifugal force by taking the vector sum of the forces from each
mass. 𝐹⃑𝑐 = 𝐹⃑𝑐,1 + 𝐹⃑𝑐,2 + ⋯
a. Graphical method
On graph paper, draw the two centrifugal force vectors head to tail. Connect
the tail of the first vector to the head of the second vector to find the sum.
Measure this vector to get approximate values of the magnitude and
direction of 𝐹⃑𝑐 .
b. Analytical method
Add the two centrifugal force vectors by breaking them into normal
components.
𝐹𝑥 = 𝐹 cos 𝜃 , 𝐹𝑦 = 𝐹 sin 𝜃
𝐹𝑦 (2)
𝐹 = √𝐹𝑥2 + 𝐹𝑦2 , tan 𝜃 =
𝐹𝑥

3. The required balancing force will be equal to the resultant force but in the opposite
direction. Therefore, add 180° to the angle obtained.

4. Now find the magnitude of the balancing mass, such that

𝐹𝑏𝑎𝑙 = 𝑚𝑏 𝑟𝑏 𝜔2 (3)
where 𝑚𝑏 = balancing mass, 𝑟𝑏 = radius of rotation (2.25” or 2.75”, depending on
location), and 𝜔 = angular velocity.

20
 (B). CAM/FOLLOWER EXPERIMENT

Introduction
A cam-follower system (sometimes referred to as a direct contact mechanism) is
one of the simplest mechanisms used for control and conversion of one type of motion to
another. However, the accuracy of the output, which is generally the motion of follower,
depends on the cam and follower being in contact at all times.

The linear motion of a follower tracing a cam profile can be defined as 𝑍(𝜃), where
𝑍 = Vertical displacement of the follower
𝜃 = Angular displacement of the cam (rad).
A logical choice is to define 𝑍 = 0 when the follower is at its lowest position. As long as
the follower maintains contact with the cam, the forces at a given rotational speed can be
found using Equation 4. 𝛽 will equal zero in this lab, since the cam pushes the follower
directly upwards (against gravity).
𝑑2 𝑍
𝑅 − 𝑊 cos 𝛽 − 𝐹𝑆 = 𝑚𝜔2 2 (4)
𝑑𝜃

𝑅 = Cam reaction force

𝑊 = Weight of the follower, retainer, and added masses
𝛽 = Angle between the line of motion of the follower and the vertical
𝐹𝑆 = Spring force on the follower
𝑚 = Mass of the follower, retainer, and added masses
𝜔 = Angular speed of the cam (rad/s)

The spring force may be broken into two components:

𝐹𝑆 = 𝐹𝑆,0 + 𝑘𝑍
(5)
𝐹𝑆,0 = 𝑘[(𝐹𝑟𝑒𝑒 𝑙𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑠𝑝𝑟𝑖𝑛𝑔) − (𝐿𝑒𝑛𝑔𝑡ℎ 𝑤ℎ𝑒𝑛 𝑍 = 0)]

𝐹𝑆,0 = Preload (spring force when the follower is at its lowest position)
𝑘 = Spring constant

The cam reaction force 𝑅 will always be positive or zero. However, a highly
negative acceleration term 𝜔2 ⋅ 𝑑 2 𝑍/𝑑𝜃 2 may require a negative value of 𝑅 in Equation 4.
In this case, the follower will not be able to maintain contact. For a given cam profile, the
critical rotational speed of the camshaft that results in separation of the follower from the
cam may be calculated from Equation 6. The value of 𝑑 2 𝑍/𝑑𝜃 2 used in Equation 6 is the
largest negative value along the cam profile.
𝑟𝑎𝑑 −(𝑊 cos 𝛽 + 𝐹𝑆,0 + 𝑘𝑍)
𝜔𝑐𝑟 ( )=√ (6)
𝑠 𝑑2𝑍
𝑚 ( 2)
𝑑𝜃 𝑚𝑖𝑛

Obviously, the speed at which the follower separates from the cam is a function of
the weight of the follower, the spring force behind the follower, and the cam profile. Once
the follower loses contact with the cam, reestablishing contact may generate a significant

21
impact force, which in turn may eventually result in fatigue failure of the surfaces of the
cam and the follower.

Objective
In this experiment, students will…
• Observe the follower behavior and determine the effect of weight of the follower
and spring force on the critical speed 𝜔𝑐𝑟 of the cam shaft (i.e., the speed at which
the follower and cam temporarily lose contact),
• Predict critical speed using measured data and judicious choice of curve fit.

Necessary Equipment and Material

• Cams with different profiles (convex, concave, tangent)
• A roller and flat follower
• Springs with different stiffness values (see Table 1)
• Variable speed drive motor
• Necessary instrumentation and recording devices
• Experimental apparatus frame and assembly (see Figure 4)

Figure 4. Cam apparatus diagram

Table 1. Spring and Retainer Specifications

Spring Weight Nominal Stiffness Retainer Weight
Color
(lb) (lbf/in.) (lb)
Red (R) 0.138 31.4 0.156*
White (W) 0.294 22.5 0.300*
Black (B) 0.156 19.7 0.144*
* Steel retainer

22
 Table 2. Additional Equipment Specifications
Weight of roller follower and attachments 3.78 lb (excluding spring, retainers, and
additional weights)
Weight of flat follower and attachments 4.30 lb (excluding spring, retainers, and
additional weights)
Diameter of roller follower 1 1/8 in.
Diameter of paper recording drum 3.673 in.

WARNING!
This lab involves high-speed rotating equipment. Loose clothing, loose jewelry,
etc. may not be worn while operating the equipment! Long hair must be properly
secured to prevent the possibility of becoming entangled in the machinery.

WARNING!
This lab involves loud sounds. To prevent hearing damage, all persons in the
vicinity must wear hearing protection! Hearing protection should not be removed
while the equipment is running, except for brief periods when necessary to detect
separation of the cam and follower.

Procedure
1. Select the direction of rotation of the cam by turning the switch provided for this
purpose. Do not change the direction of rotation while the motor is running, as this
will overload the motor.

2. Note the shape of the cam on the camshaft. Do not remove the cam from the
shaft. Cams should only be changed by the lab technicians.

3. Select one of the followers and assemble it on the machine (note that if the cam
has any concave section the follower must be a roller follower).

4. Wrap a piece of graphing paper around the recording drum and secure its ends
with scotch tape. Loosen the bolts holding the recording drum assembly in place
and connect a timing belt between the drum pulley and the camshaft pulley. Pull
the belt taut and re-secure the bolts. Turning the camshaft by hand, trace the
displacement of the follower on the paper. You will use this displacement curve
later in the calculations. Once you have traced a satisfactory curve, remove the
paper from the drum. Remove the timing belt from the drum before turning the
motor on.

5. Select one of the springs and place it in the machine. Ensure the correct two
retainers are used with the corresponding spring. Note: After changing the spring,

23
 DO NOT overtighten the bolts that secure the retaining bar! Use gentle pressure
to ensure the bolts are snug, but can be easily removed by the next group.

6. There are two retaining nuts located below the lower spring retainer. Adjust the
top nut to apply some preload to the spring. Adjust the bottom nut to secure the
first in place. Nuts should be screwed hand-tight, not using tools!

7. Increase the speed of the motor slowly until you detect the tapping noise, indicating
the impact between cam and follower. Decrease and then increase the speed
several times to make sure that you are detecting the tapping noise the very
moment that it starts. Read the speed of the camshaft using a tachometer and
record it. This is the experimental value of the speed, 𝜔𝑐𝑟 , at which the cam and
follower separate. Because the collection of data is dependent on each individual’s
hearing sensitivity, have three people perform this step independently, recording
the values for each person.

8. Repeat step 7 an additional five times, each time adding a mass of 400 grams to
the follower. Use the available washer(s) and retainer nut (hand tighten) to ensure
the added masses do not rattle during operation.

9. Ensure you have all measurements needed to calculate critical speed (Eq. 6).

10. Change the spring force by replacing the spring with one of the other two springs.
Be sure to keep top and bottom retainers with the correct springs. Repeat steps 6
– 8 with the new spring.

11. Calculate the critical speed 𝜔𝑐𝑟 analytically using Eq.6. To obtain the value of
(𝑑2 𝑍/𝑑𝜃 2 )𝑚𝑖𝑛 , you will need to take careful measurements of the curve you
measured in Step 5. Using the software of your choice, obtain a curve fit of the
measured points to approximate the shape of the displacement curve. Generate
plots of 𝑍, 𝑑𝑍/𝑑𝜃, and 𝑑2 𝑍/𝑑𝜃 2 versus 𝜃. Use the most negative value of 𝑑 2 𝑍/𝑑𝜃 2
for your calculations in Eq. 6. Note: A judicious choice of curve fit may greatly
impact the quality of your calculations. Your 𝑍 − 𝜃 plot should also 95% confidence
intervals of the curve fit, and the report should discuss what form of curve fit was
used and why it was selected.

10. Plot the variation of 𝜔𝑐𝑟 versus follower weight for both springs. Be sure to indicate
the mean and standard deviation (or precision interval) of each data point on
graphs containing measured values.

11. Compare the theoretical values of 𝜔𝑐𝑟 with the experimental values obtained and
comment on the results.

Summary of operating conditions

2 Springs
0 g, 400 g, 800 g, 1200 g, 1600 g, 2000 g added mass per spring
Person 1, Person 2, Person 3 determine critical speed
= 36 critical speed measurements
24
 (C). SECOND-ORDER SYSTEMS EXPERIMENT

Introduction
Any system whose behavior can be described in terms of a variable and its first
and second derivatives is a second-order system. A common example is the mass-
spring-damper system, where for 1-dimensional behavior, the position 𝑥 of a mass fixed
to a stationary object by a linear spring and linear damper satisfies
𝑚𝑥̈ + 𝑐𝑥̇ + 𝑘𝑥 = 𝐹(𝑡) (7)
where 𝑚 is the mass, 𝑐 is the damping coefficient, 𝑘 is the spring constant, and 𝐹(𝑡) is an
externally applied force. By observing the dynamic behavior of a second-order system,
it is possible to indirectly determine these physical characteristics.
Damping is the existence of a force (or “effort”) that tends to oppose motion.
Viscous drag at low Reynolds numbers (e.g., Stokes flow) is proportional to the velocity,
but in the opposite direction. A dashpot is a device in which relative motion of two
components requires a fluid such as oil or air to flow through a narrow orifice. This
produces the same type of linear resistance. Non-linear damping is also common. For
example, aerodynamic drag at large Reynolds numbers is typically proportional to the
square of the velocity. Sliding friction produces a force whose magnitude depends on the
coefficient of friction 𝜇 and the normal force 𝐹𝑁 , but which acts opposite to the direction
of motion.
𝐹𝑠𝑙𝑖𝑑𝑖𝑛𝑔 𝑓𝑟𝑖𝑐𝑡𝑖𝑜𝑛 = −𝜇 𝐹𝑁 𝑠𝑔𝑛(𝑥̇ ) (8)
In many cases, even if non-linear damping exists, it can be modelled as a linear damper
for small perturbations.
Springs produce forces that oppose displacement from a neutral position. This
includes not only things like wire, coil, and leaf springs, but elastic stress in materials and
the torque from the weight of the bob on a simple pendulum. Non-linear spring forces
also exist, such as from the stress in materials undergoing plastic deformation, or in
beams undergoing large deflection.

Second-Order System Characteristics

A more general form of the second-order system of equations is
1 2𝜁 1
2
𝑥̈ + 𝑥̇ + 𝑥 = 𝐹(𝑡) (9)
𝜔𝑛 𝜔𝑛 𝑘
where, 𝜔𝑛 is the natural frequency and 𝜁 is the damping ratio of the system. For the
mass-spring-damper system, these values can be found as
𝑘 𝑐
𝜔𝑛 = √ , 𝜁= . (10)
𝑚 2√𝑚𝑘

Second-order systems are typically categorized by the damping ratio, as this

determines their behavior. Systems can be underdamped, overdamped, or critically
damped. In underdamped systems, the output 𝑥 will exhibit decaying oscillations in
response to a step input 𝐹(𝑡). These oscillations occur at the so-called damped
frequency, or ringing frequency,
𝜔𝑑 = 𝜔𝑛 √1 − 𝜁 2 . (11)

25
 The rate of decay of the amplitude of these oscillations also depends on the
damping ratio, as shown in Equation 12. Here, 𝑥0 is the amplitude of the position at an
2𝜋
arbitrary time 𝑡0 , and 𝑥𝑛 is the amplitude 𝑛 cycles later (time 𝑡0 + 𝑛 𝜔 ), where 𝑛 can be
𝑑
any integer.
𝜁 1 𝑥0
= ln ( ) (12)
√1 − 𝜁 2 2𝜋𝑛 𝑥𝑛

This is most easily determined by measuring 𝑥0 at the “peak” of an oscillation, and

measuring 𝑥𝑛 at the 𝑛-th peak after 𝑥0 . Note that if 𝜁 is sufficiently small, Equation 12 can
be closely approximated by
𝜁 1 𝑥0
≈𝜁≈ ln ( ) , 𝑓𝑜𝑟 𝜁 ≪ 1 (13)
√1 − 𝜁 2 2𝜋𝑛 𝑥𝑛

Objective
In this experiment, students will measure the dynamic output of an oscillating
system and use it to mathematically derive properties of the physical components used.

Necessary Equipment
• Model 210a Rectilinear Dynamic System.
• High-stiffness, medium-stiffness and low-stiffness springs.
• Brass masses.
• Allen wrenches.
• Computer with data acquisition board for recording and plot generation of the
Rectilinear Dynamic System.
• Flash drive for saving data files.

Figure 5. Model 210a Rectilinear Dynamic System

26
 NOTE: Data Collection
In this experiment, the raw data will be collected electronically and saved in text
files for later analysis. However, you should also create a separate raw data sheet by
doing the following. For each of the seven cases, once you have the plot of position
vs. time:
o Choose the two points 𝑥0 and 𝑥𝑛 that you will use for the analysis.
o Record approximate values for 𝑡0 , 𝑥0 , 𝑡𝑛 , and 𝑥𝑛 , plus the number of cycles in-
between (i.e., 𝑛).

You will still use the saved text files to get more precise values for the post-lab
analysis. However, this will give you a raw data sheet for your TA to sign. In addition,
it will allow you to approximate the frequency and damping ratio of each case before
the formal analysis. This way, you can check your numbers and, hopefully, catch any
calculation errors before submitting the report.

Procedure
1. With computer powered up, launch the EDyn32 software, enter the Driving
Function box via the Setup menu and select Force (Torque), then select Setup
Driving Function, then OK (from within the Force (Torque) dialog box), then
Enable Driving Function, and finally OK again to return the background screen.

2. Enter the Command menu, go to Input Shape, and select Step Input. Input a
Step size of 0 (zero), dwell time of 3000 ms and 1 repetition. Exit to background
screen by consecutively selecting OK. This puts the controller board in a mode for
acquiring 6 seconds of data on command but without actually introducing drive
force (via the drive motor).

3. Go to Data Menu - Setup Data Acquisition in the Data menu, select Encoder #1
and Encoder #2 as data to acquire, and specify data sampling every two servo
cycles, (i.e., every 2 ⋅ 𝑇𝑆 ). Select OK to exit.

CASE 1
4. Securely clamp carriage #2 in place using the stop bumpers as shown in Figure 6.
Place a ¼ in. threaded nut between each bumper and the cart so as not to engage
the limit switches on the bumpers. See that the centerline mark of carriage #1
coincides approximately with the 0 of the scale provided along carriage #1.
Move the stop bumpers for carriage #1 to the extreme outer positions, giving the
carriage room to oscillate freely. Fix the medium-stiffness spring between carriage
#1 and carriage #2. When loosening the stop bumpers, take care not to let the
nuts underneath the apparatus fall off the screws. If they do, you will have to screw
them back on.

27
 Stop bumpers

Limit switch

Carriage #2

Figure 6. Detailed view of carriage locked in place by stop bumpers

5. Secure four 500g masses on carriage #1.

6. In the Utility menu select Zero position to zero the initial encoder readings.

7. Select Execute from the Command menu. Manually displace carriage #1

approximately 2.5 cm. Exercise caution in displacing the carriage so as not to
engage the limit switch. With the first mass held at 2.5 cm from the initial position,
select Run from the Execute box and then release the mass approximately 1
second later. The mass will oscillate and attenuate while encoder data is collected
to record this response. Select OK after data is uploaded.

Note: If at any time during this procedure a limit switch is engaged, you must
return to the Driving Function box and Enable Driving Function before
proceeding.

8. Select Set-up Plot from the Plotting menu and add Encoder #1 position to the
left axis. Then select Plot data from the Plotting menu. You will see the first
carriage’s time response (position vs. time data). It should somewhat resemble
Figure 7. Select your starting and ending peaks for analysis and record them on
your raw data sheet. Note that the digital encoder outputs position as an integer
value in “counts.” The conversion factor is (2266 counts = 1 cm).

9. In the Data menu, select Export raw data and save the data in a .txt format. Open
the exported file in Wordpad or Excel to ensure it saved correctly.

28
 3

2.5

1.5
Steady State
Encoder Position (cm)

1 Error

0.5

-0.5

-1 X0 Xn

-1.5

-2 t0
tn
-2.5
0 0.5 1 1.5 2
Time (s)

Figure 7. Typical step response

10. Calculate the damped frequency 𝜔𝑑,1 of the system in Case 1.

CASE 2
11. Remove the extra mass from carriage #1.

12. In the Utility menu, zero the carriage encoder position.

13. Repeat steps 7 – 10 for the current case.

CASE 3
14. Unclamp carriage #2 and clamp carriage #1 using the same procedure as in
step 4. Ensure the centerline mark of carriage #2 coincides approximately with 0
on the provided scale. Move the stop bumpers to the outside position to give
carriage #2 room to oscillate. If the dashpot is connected to carriage #2, disconnect
it.

15. Secure four 500 g masses on carriage #2.

16. In the Utility menu select Zero position to zero the initial encoder readings.

17. Select Execute from the Command menu. Manually displace carriage #2
approximately 2.5 cm. Exercise caution in displacing the carriage so as not to
engage the limit switch. With the first mass held at 2.5 cm from the initial position,
select Run from the Execute box and then release the mass approximately 1
second later. The mass will oscillate and attenuate while encoder data is collected
to record this response. Select OK after data is uploaded.

29
 18. Select Set-up Plot from the Plotting menu, remove Encoder #1 position, add
Encoder #2 position to the left axis, and then select Plot data from the Plotting
menu. You will see the time response of carriage #2.

19. In the Data menu, select Export raw data and save the data in a .txt format.

20. Calculate the damped frequency 𝜔𝑑,3 of the system in Case 3.

CASE 4
21. Remove the extra mass from carriage #2.

22. Repeat steps 16 – 20 for the current case.

CASE 5
23. Connect the dashpot to carriage #2.

24. Set the damping adjustment knob to its fully closed position (i.e., when it first
begins to resist tightening). Do not over-tighten. You may make a reference mark
on the knob at this point.

25. Open the damping adjustment knob 2 complete turns (720°) from its fully closed
position.

26. Secure four 500 g masses on carriage #2.

27. Repeat steps 16 – 20 for the current case.

CASE 6

28. Disconnect the dashpot and remove the extra mass from carriage #2.

29. Replace the medium stiffness spring with a high-stiffness spring.

30. Repeat steps 16 – 20 for the current case (high-stiffness spring, no added mass).

CASE 7
31. Replace the high-stiffness spring with the low-stiffness spring.

32. Repeat steps 16 – 20 for the current case.

33. When you have finished, shut down the software. Ensure that you have all seven
data files copied to two secure locations, then delete them from the PC.

30
Calculations
1. Generate a plot of displacement (in cm) versus time for each of the seven cases.

2. For each of the seven cases, calculate:

a. The damped frequency 𝜔𝑑,𝑖
b. The damping ratio 𝜁𝑖
c. The natural frequency 𝜔𝑛,𝑖
Note: In some cases (due to high friction, slip of the encoder wire, etc.), there may
be a significant steady-state error. That is, the encoder position may settle towards
a value other than zero, as shown in Figure 7. If any case has a steady-state error
greater than 0.05 cm, you should subtract the error from your 𝑥0 and 𝑥𝑛 values
when calculating the damping ratio.

3. Calculate the following quantities:

𝑚𝐶1 Mass of carriage #1

𝑚𝐶2 Mass of carriage #2

𝑘𝐻 Spring constant of high-stiffness spring

𝑘𝑀 Spring constant of medium-stiffness spring

𝑘𝐿 Spring constant of low-stiffness spring

𝑐𝐶1 Damping coefficient of carriage #1

𝑐𝐶2 Damping coefficient of carriage #2

𝑐𝐷 Damping coefficient of carriage #2 with dashpot attached

To do this, use Equations 10 for each of the seven cases. For example:
Case 1: 𝑚1 = 𝑚𝐶1 + 𝑚𝑎𝑑𝑑𝑒𝑑
𝑘1 = 𝑘𝑀 𝑘1 𝑘2
𝜔𝑛,1 = √ , 𝜔𝑛,2 = √
Case 2: 𝑚2 = 𝑚𝐶1 𝑚1 𝑚2
𝑘2 = 𝑘𝑀

This gives you a set of two equations, from which you can calculate the two
unknowns 𝑚𝐶1 and 𝑘𝑀 .

Proceed similarly to solve for the other quantities.

Note that you may be able to solve for some quantities in more than one way. You
are encouraged to do so, compare your answers, and comment on them.

31
[This page intentionally left blank.]

32
 (D). STRAIN GAUGE EXPERIMENT

Introduction
The design of the systems that involve components with complicated geometry,
complicated loading and perhaps points of stress concentration, are often verified by
actual measurement of strains, from which the actual stresses can be calculated. A
common technique used in practice is the application of strain gauges. The utilization of
a strain gauge is comparatively easier and less expensive than methods such as
photoelasticity. It measures the strain only at the surface of the object. In order to find
the three unknowns surface strains 𝜀𝑥 , 𝜀𝑦 , and 𝛾𝑥𝑦 at any point on an 𝑥𝑦 plane, it is
necessary to use three gauges oriented in three arbitrary directions. However, if the
direction of principal stresses is known, then two gauges are sufficient if they are installed
parallel to the principal axes. These principal stress directions can be easily determined
by using the stress coat application method. The stress coat is a brittle lacquer that is
applied on the surface in liquid form (normally by spraying). When dried, in about 3-4
hours, it will become brittle. The coating will crack perpendicular to the direction of tensile
principal stresses if the member is subjected to stress and if the strain on the surface is
above the coating’s threshold strain. Observing the cracks, one can determine the
direction of principal stresses.

Principle of Strain Gauges

Strain gauge measurement is based on change of electrical resistance of a
conductor. The electrical resistance of a wire is determined by
𝜌𝐿 ⬚
𝑅= (14)
𝐴 ⬚
where 𝜌 is the resistivity, 𝐿 is the length, and 𝐴 is the cross-sectional area of the wire.
When a strain is applied to the surface, the length of the wire will change as well as its
cross-sectional area (due to Poisson’s effect) and its resistivity (due to the piezoresistive
effect). This causes a change Δ𝑅 in the resistance of the wire. This change is related to
the linear surface strain as
Δ𝐿 Δ𝑅/𝑅
=𝜀= (15)
𝐿 𝑓
where 𝑓 is a gauge factor provided by the manufacturer. To reduce the footprint of the
gauge and more accurately approximate the strain at a point, the conductor is typically
formed or stamped out of a thin metal foil in the boustrophedon configuration shown in
Figure 8. Commercial gauges are available in a variety of sizes, with some applications
requiring gauges smaller than 1/64 of an inch in length and width. A strain gauge may
contain only one sensing element, which is called single-element gauge, or it may be a
rosette containing two or three independent sensors set at fixed angles relative to one
another.
A Wheatstone bridge as shown in Figure 9 measures the change of electrical
resistance in the gauge due to strain. The instrument that does this is called strain
indicator. Once the gauge is mounted on a surface, it is connected to the bridge and the
circuit is balanced (zero potential between B and D) before the member is loaded. This
is accomplished by changing the resistance of other legs until the indicator reads zero.
When the strain is applied, the bridge will have to be rebalanced. The strain indicator is

33
 Figure 8. Typical strain gauge configuration Figure 9. Wheatstone bridge with
strain gauge and galvanometer

calibrated using the gauge factor 𝑓 such that the change in resistance required to
rebalance the bridge correlates to the strain measured by the strain gauge. This is known
as the null method. Strain indicators will typically display the strain directly in units of
microstrain, 𝜇𝜀 (i.e., micro-inch per inch, 𝜇𝑖𝑛/𝑖𝑛).

Stresses in a Cantilever Beam

A round beam supporting a weight 𝑃 off center as shown in Figure 10 will experience a
bending moment 𝑀 and a torque 𝑇. The stress distribution at a point on the beam can
be predicted using static beam theory, or measured experimentally using a three-element
strain gauge rosette.

If the longitudinal, tangential, and shear strains 𝜀𝐿 , 𝜀𝑇 , 𝛾𝐿𝑇 on a beam are known,
the linear strain at any angle 𝜃 from the longitudinal direction can be found as
𝜀𝐿 + 𝜀𝑇 𝜀𝐿 − 𝜀𝑇 𝛾𝐿𝑇
𝜀𝜃 = + cos(2𝜃) + sin(2𝜃) ⬚ (16)
2 2 2 ⬚

Figure 10. Round beam under a bending moment and a torque. A strain gauge is
mounted on top of the beam.

Conversely, if we measure the strain at three given angles using strain gauges, we can
substitute these into Equation 16 to obtain three equations and solve for 𝜀𝐿 , 𝜀𝑇 , and 𝛾𝐿𝑇 .
We can then find the principal strains 𝜀1 and 𝜀2 using Equation 17.
34
 2
𝜀𝐿 + 𝜀𝑇 √(𝜀𝐿 − 𝜀𝑇 )2 + 𝛾𝐿𝑇
𝜀1,2 = ± (17)
2 2

The principal stresses 𝜎1,2 and the direction 𝛼 of the first principal stress can be
found using Equation 18, where 𝐸 is the modulus of elasticity and 𝜈 is the Poisson ratio.
𝐸
𝜎1 = [𝜀 + 𝜈𝜀2 ]
1 − 𝜈2 1
𝐸
𝜎2 = [𝜀 + 𝜈𝜀1 ] (18)
1 − 𝜈2 2
𝛾𝐿𝑇
tan(2𝛼) =
𝜀𝐿 − 𝜀𝑇

The principal stresses at any point on the surface of the beam can be predicted
theoretically using Equation 19, where 𝐼 is the area moment of inertia of the section, 𝐽 is
the polar moment of inertia, ℎ is the vertical distance from the neutral axis to the point
where the strain is measured, and 𝑟 is the radius of the beam. This is equivalent to
constructing the Mohr’s circle.
𝜎𝐿 + 𝜎𝑇 𝜎𝐿 − 𝜎𝑇 2 2
𝜎1,2 = ± √( ) + 𝜏𝐿𝑇
2 2
2𝜏𝐿𝑇 (19)
tan(2𝛼) =
𝜎𝐿 − 𝜎𝑇
𝑀ℎ 𝑇𝑟
𝜎𝐿 = , 𝜎𝑇 = 0, 𝜏𝐿𝑇 =
𝐼 𝐽

Objective
In this experiment, students will…
• Use a strain gauge rosette and strain indicator to measure strain;
• Calculate stress on a part using strain gauge measurements and compare the
results with theory;
• Investigate the effect of strain gauge alignment on their results;
• Gain hands-on experience mounting a strain gauge to a metal sample.

35
Necessary Equipment and Tools
• Aluminum bar specimen
• Strain gauge rosette (𝑓 = 2.04)
• Pre-wired strain gauge for application (𝑓 = 2.08 ± 1.0 %)
• Specimen preparation and strain gauge application supplies
• Switch and balance unit with strain indicator
• Ohmmeter (or multimeter)
• Means to apply load to the specimen
• Specimen load (weights)

Procedure
Part A – Strain Gauge Application
1. Your TA will provide you with a flat bar specimen and a pre-wired linear strain
gauge. You will be installing the strain gauge on this bar at a distance of 2.5” from
one end. Identify one such area that does not have a strain gauge currently
installed. Using the metal ruler, make two marks (approx. ¼” long) on either side
of the bar that are 2.50” from the end of the bar (see Fig. 11).

Figure 11. Marking the placement of your strain gauge on the flat bar

2. Using a permanent marker, label the bar with your section and group number (e.g.,
S1G2) about an inch from the area where your strain gauge will be.

3. Take your bar and strain gauge to the application area. Follow the instructions in
the Strain Gauge Application Supplemental Document to glue the strain gauge to
your beam in the location indicated. Orient your strain gauge so that the wire leads
point towards the center of the beam, as shown in Figure 12.

Figure 12. Strain gauge orientation on flat bar

4. After applying your strain gauge, move on to Part B. This will allow time for the
adhesive to cure before you attempt to take measurements.

36
Part B – Strain Gauge Rosette
1. Examine the round aluminum bar provided. Carefully examine and record the
construction and orientation of the gauges. Do not touch the strain gauges or any
of the wires connected to them to avoid damage!

2. Make sure there is a good connection between the wire leads and the strain gauge
by measuring the resistance across each gauge (at the test ports) with an
ohmmeter. Record the resistance values. (Ensure the switch and balance unit is
turned off, as it will interfere with the ohmmeter measurement.)

3. Connect the gauges to the strain indicator through a switch and balance unit (if not
already connected), and balance the circuit. Note what type of Wheatstone bridge
configuration is used.

4. Take all necessary physical measurements to calculate theoretical stress at the

location of the strain gauge rosette.

5. Apply load to the specimen in increments of 2 lb up to 20 lb. At each step, record

the load and the three measured strains.

6. Remove the load, re-balance the strain indicator, and repeat steps 3 and 4 two
more times, for a total of three data sets.

Part C – Testing Specimen

1. Recover the flat specimen from Part A and remove the adhesive tape. Ensure that
the strain gauge is securely glued to the surface. Gently touch the ohmmeter
probes to the free ends of the strain gauge wires to measure the resistance of the
strain gauge and ensure they were not broken during application.

2. Mount the specimen in the bending assembly vise, with your strain gauge mounted
on top and near the vise grip. Ensure the back edge of the specimen is flush with
the stop provided. Lock the specimen securely in place.

3. To connect your strain gauge to the switch and balance unit, open each of the
quick connects provided and insert one of the wire leads from your strain gauge.

4. Take all necessary physical measurements, including the dimensions of the bar
and the locations of the strain gauge and the applied load.

5. Pass the threaded rod of the weight support through the hole in your bar and
secure it in place using the nuts and washers provided.

6. Balance the strain indicator, then apply a 2 lb load to the specimen and wait for it
to settle. Record the measured strain, then remove the load.

7. Repeat step 6 until you have a total of five strain measurements.

37
 8. Remove your specimen from the vise and turn off the switch and balance unit.

Analysis
For the results of Part B:
1. Provide a linear regression for the load vs. strain data for each of the three strain
gauges. Show the error bar of each load based on the 95 % precision interval of
the three measurements.

2. Using a theoretical load of 𝑃 = 15 𝑙𝑏𝑓 and your curve fits from step 1, calculate and
report 𝜀𝐿 , 𝜀𝑇 , 𝛾𝐿𝑇 ; 𝜀1 , 𝜀2 ; 𝜎1 , 𝜎2 , and 𝛼 using Equations 16 – 18.

3. Using your results from step 2, calculate a sensitivity index 𝜕𝜀𝜃 /𝜕𝜃 for each of the
three strain gauges (convert the units to microstrain per degree) and comment on
the results. What do these numbers physically represent?
(Hint: Review propagation of uncertainty.)

4. Using a theoretical load of 𝑃 = 15 𝑙𝑏𝑓 , calculate and report theoretical values for
𝜎1 , 𝜎2 , and 𝛼 using Equation 19.

5. Compare the results from step 2 and step 4.

6. Provide additional discussion, including the maximum load 𝑃𝑚𝑎𝑥 that the aluminum
bar could safely support.

For the results of Part C:

7. Calculate the mean measured strain and its 95 % confidence interval.

8. Calculate the theoretical bending strain.

9. Compare the results from step 7 and step 8.

38
 (E). JOURNAL BEARING LUBRICATION EXPERIMENT

Introduction
The major objective of lubrication in journal bearings is to induce and maintain a
film of lubricant between the journal and the bearing. The purpose of this film is to keep
the two surfaces separate at all times and thus prevent dry contact which otherwise can
lead to bearing failure.

Hydrodynamic lubrication is the most

common method of lubrication for journal bearings.
In this method, as the shaft rotates, due to the load
applied to it (as well as its own weight), it will take a
slightly eccentric position relative to the bearing. The
eccentric rotation of the shaft in the bearing, as
shown in Figure 13, acts somewhat like a rotary
pump and generates a relatively high hydrodynamic
pressure in the converging zone. The hydrodynamic
pressure for a properly designed bearing is
responsible for supporting the shaft without allowing
it to come in contact with the bearing. A stable oil
film will occur when the ratio 𝜇𝑁/𝑃, is above a critical Figure 13. Eccentric shaft
value. Below this value, the oil film is unstable, and rotating clockwise in bearing
the journal will begin making contact with the bearing
(“rub”). The upper limit of operating speeds is often determined by a phenomenon known
as oil whirl. Oil whirl is a type of fluid-induced instability in the flow between the journal
and the bearing. It can cause the journal to orbit around the center of the bearing, as well
as produce significant vibrations. A number of journal bearing designs have been
proposed to mitigate the problem of oil whirl, which you are encouraged to research on
your own.

The hydrodynamic pressure distribution around an infinitely long, fully lubricated

bearing can be derived analytically, and is given by
(2 + 𝜀 cos 𝜃) sin 𝜃
(𝑝 − 𝑝0 ) = 𝐾 [ ] (20)
(1 + 𝜀 cos 𝜃)2
where,
12𝜋𝜇𝜀𝑟 2 𝑁
𝐾= 2 (21)
𝑐 (2 + 𝜀 2 )
The remaining terms in the above equations are defined as,
𝑟 = Journal radius (in)
𝑐 = Mean radial clearance (in)
𝜇 = Oil viscosity at operating temperature (reyn)
𝑒 = Eccentricity (in)
𝜀 = Eccentricity ratio (= 𝑒/𝑐)
𝜃 = Angle with respect to line of centers (rad)

39
 𝑁 = Journal speed (RPS)1
𝑝0 = Mean oil pressure in bearing (psi)
𝑝 = Hydrodynamic pressure at position 𝜃 (psi)

The location of maximum pressure is denoted by the angle 𝜃1 .

(𝑝 − 𝑝0 ) = (𝑝 − 𝑝0 )𝑚𝑎𝑥 𝑎𝑡 𝜃 = 𝜃1 (22)
Using this definition and Equation 20, the relation between the location of maximum
pressure and the eccentricity ratio, shown in Equation 23, can be derived analytically.
The minimum film thickness, ℎ0 , relates directly to eccentricity, as shown in Equation 24.
3𝜀
cos 𝜃1 = − (23)
2 + 𝜀2
ℎ0 + 𝑒 = 𝑐 (24)
A dimensionless number called the Sommerfeld Number or bearing characteristic
number, 𝑆, relates the bearing performance to the design parameters: It is defined as
𝑟 2 𝜇𝑁
𝑆=( ) ( ) (25)
𝑐 𝑃
where,
⬚𝑃 = 𝑊 ⬚ (26)
⬚ 2𝑟𝐿 ⬚
𝑃 = Load per projected area of the bearing (psi),
𝐿 = Bearing length (in),
𝑊 = Load (weight) carried by the bearing (lb).
𝑟
The clearance ratio ℎ0 /𝑐 and the coefficient of friction variable 𝑐 𝑓 are plotted
versus Sommerfeld number in Figures 14 and 15, respectively. The latter is used to find
the coefficient of friction, 𝑓. This is related to the frictional torque, 𝑇, on the bearing.
𝑇 = 𝑓𝑊𝑟 (27)
In actual journal bearings, the pressure is not constant along the length of the
bearing. The differential pressure magnitudes will be reduced towards either end of the
bearing, much like lift is reduced near the wingtips of an aircraft. This affects the loading,
as can be seen by the effect of the ratio 𝑙/𝑑 in Figures 14 and 15. In this experiment, you
will account for the axial pressure distribution by finding a pressure ratio, 𝑅, and applying
it at all values of 𝜃.
(𝑝(𝜃) − 𝑝0 )𝑎𝑣𝑒𝑟𝑎𝑔𝑒
𝑅= | (28)
(𝑝(𝜃) − 𝑝0 )𝑚𝑎𝑥 𝜃=𝑐𝑜𝑛𝑠𝑡.

Objective
In this experiment, students will…
• Measure hydrodynamic pressure variation in a journal bearing at different speeds
and at different loads;
• Calculate hydrodynamic lift on the bearing and compare it with the actual load;
• Predict bearing eccentricity using the measured pressure profile;
• Calculate the friction losses in the bearing.

1 RPS denotes revolutions per second or rev/s.

40
Necessary Equipment and Material
• Journal bearing with adjustable speed journal
• Instrumentation to measure the pressure around the bearings (manometer bank)
• Instrumentation to measure journal RPM (tachometer)
• Dead weights to adjust the load on the bearing

Procedure
1. Before starting the motor, record the initial (static) head in the bearing, 𝐻𝑆 . If the
manometer tubes are not even, take the average of all 16 taps and use this value.
This is for comparison against the average pressure while running.

2. Select the direction of rotation of the motor by setting the switch provided for this
purpose. Do not change the direction of rotation while the motor is running.

3. A load should already be applied on the bearing using the dead weights (if not, tell
your TA). Record the amount of the load added.

4. Start the motor and let it run for about half an hour, or until the temperature and
the oil viscosity reach a steady state condition. Visually verify the direction of
rotation of the shaft.

5. Set the journal speed to 1200 RPM. (Be sure to check this speed periodically as
it may increase during the course of the experiment.) Allow enough time for the
oil to level in the manometer tubes to stabilize. (Manometer readings taken two
minutes apart should give the same value for the levels to be stabilized.)

6. Read the pressures for locations 1 through 16 and convert the readings to psi.

7. Note the position of the shaft in the bearing.

8. Change the shaft speed to 1600 RPM and repeat steps 5 through 7 for this speed.

9. Gradually increase the shaft speed until you observe signs of oil whirl. Record the
approximate critical speed, 𝑁𝑐𝑟 , at which the oil whirl begins, and record the
behavior of the bearing during oil whirl.

IMPORTANT: DO NOT allow the journal to operate in oil whirl for more than
10 seconds at a time!! This is to prevent damage to the equipment.

10. Turn off the motor and ask your TA to remove the weights from the bearing. Once
this is done, start the motor again and repeat steps 5 through 7 at 1600 RPM.

11. When finished, turn off the motor and ask your TA to reattach the weights to the
bearing.

41
For Each of the Three Data Sets:
12. Plot the mid-bearing pressure (taps 3 and 6-16) vs. angle 𝛽 in Cartesian as well
as polar coordinates. (Note that 𝛽 = 0 corresponds to tap 3.)

13. Find two points, A and B, on the experimental pressure curve that are 180° apart
but have equal pressure. Note that for any pressure curve, there will be only one
such pair of points possible. These two points, A and B, form the axis 𝑝 − 𝑝0 = 0
for the Sommerfeld curve (i.e., the line of centers). Report this pressure, 𝑝0 , and
compare against the static pressure, 𝑝𝑆 (from Step 1).

14. Of these two points, choose as the origin the point with a larger thickness of oil
film, and take the axis 𝜃 = 0° to pass through this point.

15. Subtract 𝑝0 from the tap readings to convert to differential pressure (𝑝 − 𝑝0 ). Plot
the variation of differential pressure (𝑝 − 𝑝0 ) along the bearing axis (pressure taps
1-5). Obtain the average pressure along the axis as well as the ratio, 𝑅, of average
pressure to mid-bearing pressure (tap 3).

16. Multiply the differential pressures of taps 6-16 by 𝑅 to obtain axially averaged
pressure values at each value of 𝜃.

17. From your pressure distribution graph, determine the location (𝜃1 ) of maximum
pressure, then from Equation 23 find the eccentricity ratio.

18. Use Equations 20 and 22 to calculate the constant 𝐾 (report this value), then
calculate the Sommerfeld curve (theoretical pressure distribution) using Equation
20. An example curve is shown in Figures 16 and 17. Use enough points to create
a smooth (and accurate) curve.

Plot this theoretical Sommerfeld curve (using lines) and the axially averaged
pressures from Step 16 on the same graph, and compare the two. Be sure the
graph indicates whether these are plotted against 𝛽 or against 𝜃 (these are not the
same).

19. Calculate the theoretical load on the bearing using Equation 29.
12
𝜋𝐿𝑑
𝑊𝑡ℎ𝑒𝑜𝑟. = ∑ ( ) (𝑝 − 𝑝0 )𝑖 cos 𝛽𝑖 (29)
12
𝑖=1
Here, (𝑝 − 𝑝0 )𝑖 is the pressure at any pressure tap location and 𝛽𝑖 is the angle
between that location and the line of action of load (i.e., 𝛽 = 0 at tap point number
3). Compare this load with the actual load on the bearing and discuss.

20. Calculate the Sommerfeld number from Equation 30.

2
⬚𝑆 = 𝐾(2 + 𝜀 ) ⬚ (30)
⬚ 12𝜋𝜀𝑃 ⬚
Then, find the friction coefficient, 𝑓, using Figure 15.
42
21. Calculate and report the theoretical power (in horsepower) lost to bearing friction
versus rotational speed for the three cases.

22. Calculate the lubricant viscosity at each case using the Sommerfeld number.

23. Discuss your results and comment on them.

43
Figure 14. Variation of 𝒉𝟎 /𝒄 with Sommerfeld Number

Figure 15. Variation of (𝒓/𝒄)𝒇 with Sommerfeld number

44
Figure 16. Comparison of theoretical and experimental pressure curves

45
Figure 17. Polar pressure diagram

46
 (F). VIBRATION ANALYSIS OF ROTATING EQUIPMENT EXPERIMENT

Introduction
Rotating equipment is used in a wide variety of processes, including chemical
processing plants, power generation plants, and the common automobile. Some
examples of rotating equipment include electric motors, pumps, compressors, turbines,
generators, alternators, gearboxes, and transmissions. Often, one piece of rotating
equipment must be coupled with a driving rotating component, such as a water pump
being driven by an electric motor. Distinct and possibly damaging vibrations are
generated when coupled equipment is imbalanced, misaligned, or loose [1]. This
experiment will focus on measuring these vibrations and using their frequency profile to
identify these types of faults in rotating equipment.
In this lab, vibrations in rotating machinery will be measured using an
accelerometer. The rotating machinery has been specially designed to introduce
measurable faults that mimic some of those frequently seen in practice. In order to
identify the fault, it is necessary to look at the signal in the frequency domain. This is
done by running the acceleration signal through a Fast Fourier Transform (FFT). The
resulting frequency spectrum gives us details about the frequencies that are present [2].
In vibration analysis, the running speed (shaft speed) of the rotating equipment is
commonly denoted as the “1× peak” or the turning speed (TS), and frequently appears
as the fundamental frequency of the dominant vibrations in rotating machinery. Multiples
of this frequency, known as the harmonic frequencies, may be denoted as 2×, 3×, 4×,
and so forth. For example, a pump driven by an electric motor at 1800 RPM (30 Hz) will
have a 1× peak at 30 Hz on a graph of the frequency spectrum. A peak at 3600 RPM (60
Hz) on this spectrum would be denoted as the 2× peak, or the second harmonic.

Fault Type: Imbalance

Imbalance occurs in rotating equipment when the axis of rotation does not coincide
with one of the principal axes of the rotating object. There are two types of imbalance.
An object is statically unbalanced if its center of mass does not lie on the axis of rotation.
An object is dynamically unbalanced if, while rotating, centrifugal forces create a net force
couple on the rotating equipment. Both types of imbalance may exist simultaneously.
Static imbalance may be corrected by adding or removing mass from the rotating
equipment at a single location. Dynamic imbalance requires adding or removing mass at
multiple locations.
If an object is initially balanced (statically and dynamically), then adding an
additional mass 𝑚𝑏 at radius 𝑟𝑏 from the axis of rotation creates an imbalance 𝑚𝑏 𝑟𝑏 .
When the object spins at angular velocity 𝜔, this mass will produce a centrifugal force 𝐹𝑐
on the object.
⬚𝐹 = 𝑚 𝑟 𝜔 2 ⬚ (31)
⬚ 𝑐 𝑏 𝑏
⬚
While the magnitude of this force is constant, its direction changes as the shaft
rotates. Thus, if we have an accelerometer mounted to a stationary bearing, the
centrifugal force will be aligned with the accelerometer once per revolution. The
frequency spectrum of vibrations associated with imbalance tend to have a dominant 1×
peak (i.e., at the running speed), as shown in Figure 18. The harmonics (2×, 3×, etc.) will
have comparatively low amplitudes.
47
Figure 18. Imbalance signature spectrum, distinct 1×TS frequency peak

Fault Type: Looseness

Looseness occurs when machinery has some freedom to move in a direction for
which it was not designed. For example, vibrations may cause a bolt to become loose,
or rusting may cause a washer to fail and break away. In the case of rotating equipment,
if centrifugal forces are strong enough, a loose support may cause the bearing to move
up and down during operation. The displacement of a loose bearing tends to resemble a
square wave – the bearing will stay in contact with the support for a time, then lift off until
it impacts an upper support, then stay there a while until it falls back down and impacts
the lower support again.
In the frequency domain, this square shape will contain significant energy at
multiple harmonics of the running speed, as can be seen in Figure 19. Figure 20 shows
the spectrum from a centrifugal pump at a chemical plant in Geismar, LA. The distinct
4th, 6th, and 8th harmonic responses are a strong indicator of looseness in the pump.

Figure 19. Looseness signature spectrum, distinct 4×TS frequency peak among multiple
other significant harmonics

48
Figure 20. Velocity spectrum from Geismar plant pump, distinct 4×, 6×, and 8×TS peaks

Fault Type: Misalignment

Misalignment occurs when two shafts are coupled together, but their axes of
rotation are not collinear. Offset misalignment occurs when the centers are offset from
one another, and angular misalignment occurs when the shafts are not parallel. In
practice, a combination of the two is more likely to occur.
Misalignment typically manifests as dominant vibrations at the 2× speed, while the
1× and other peaks may also be prominent. Angular misalignment tends to increase the
amplitude of the 1× peak, while offset misalignment leads to a higher 2× peak. Figures
21 and 22 show the spectra from a misalignment module at LSU and from a blower at a
chemical plant in Geismar, LA, respectively. In both cases, the relatively high strength of
the 2nd harmonic response compared to the other harmonics is a good indicator of
misalignment.

Vibration analysis, in conjunction with visual inspections, can be used to diagnose

a wide variety of other machine faults, including belt or gear train issues, oil whirl, ball
bearing issues, and more. Additional vibration information is often required for proper
analysis. This may include measuring acceleration in multiple directions (via multiple
accelerometers or 3-axis accelerometers), measuring in multiple locations, and looking
at phase differences between signals.

49
Figure 21. Misalignment signature spectrum, 50 Hz turning speed, distinct 2×TS peak

Figure 22. Velocity spectrum from Geismar plant blower, distinct 2×TS peak suggests
misalignment

Bearing Life
Rotating machinery is supported by one or more bearings, which allow the shaft to
rotate with minimal friction. Both normal operation and fault-induced vibrations can
produce significant cyclical loads on the bearings, so they must be carefully chosen to
avoid the bearings failing prematurely due to fatigue.

50
 To estimate the life of the bearing, we first need to find the equivalent force 𝐹𝑒 that
represents the combined radial load 𝐹𝑟 and thrust load 𝐹𝑡 . If the thrust load is less than
35 % of the radial load, as it is in this lab, then we can use the radial load as the equivalent
load.
𝐹𝑡
0 < < 0.35, 𝐹𝑒 = 𝐹𝑟 (32)
𝐹𝑟
The bearing life 𝐿, in number of cycles, can be found using Equation 32.
⬚ 𝐶 3.33 ⬚
⬚𝐿 = 𝐾𝑟 𝐿𝑅 ( ) ⬚ (33)
𝐹𝑒 𝐾𝑎
⬚ ⬚
• 𝐿𝑅 = Standard life cycles
• 𝐾𝑟 = Reliability factor
• 𝐾𝑎 = Application factor
• 𝐶 = Bearing capacity

The standard life is commonly set to be 106 cycles. The bearings used in this
experiment are YAR 203-010-2F ball bearings from SKF, which have a rated capacity of
𝐶 = 9.56 𝑘𝑁. The reliability factor and application factor can be found from Figures 23
and 24, respectively.

Figure 23. Reliability factor [3]

51
Figure 24. Application factor table [3]

Objective
In this experiment, students will…
• Record accelerator data measured from a variety of simulated faults in rotating
machinery;
• Use the frequency spectra to identify the faults present;
• Calculate centrifugal and impact forces and predict bearing life under periodic
loading.

Necessary Equipment and Materials

• Vibration Analysis Modular System (Fig. 25), includes:
o Variable-frequency drive (VFD) controller
o Three AC motors on independent contactor circuits C1, C2, and C3
o Siemens programmable panel to open/close circuits
o Three stainless steel shafts, each coupled to a motor (8 inch distance between
ball bearings)
o Two 1 kg brass flywheels (diameter 5 inches, bolt holes at a 1.5 inch radius)
• Bolts for unbalancing flywheels (mass of 24.60 g ± 0.01 g)
• Acoem FALCON handheld vibration analyzer with touch-sensitive screen
• Integrated Electronics Piezo-Electric (IEPE) accelerometer with magnetic base
• PC with NEST i4.0 software installed
• Flash drive for transferring data files

NOTE: Keep lab table clear!

During this experiment, you will be measuring vibrations on the Vibration Analysis
System. The lab table is a part of this system and can influence the results at certain
frequencies. Do not lean on or bump the table, nor leave heavy objects on the table,
while you are collecting data, as these may affect your results!

52
Figure 25. Vibration Analysis Modular System showing locations of shafts and bearings.

Procedure
1. Insert the flash drive into the PC. On the flash drive, create a folder named “Export”
if it does not already exist. This will be used to export data files to the NEST
software. If the folder already exists, you may delete any files in the folder.

2. Ensure that none of the motor power circuits are on by doing the following: On the
Siemens control panel, use the arrow buttons to navigate the cursor to “C1 Off,”
then press the green “ENTER” button. Repeat for “C2 Off” and “C3 Off.”

Shaft C1

3. Remove any bolts from the brass flywheels. Place the accelerometer vertically on
the outboard bearing housing of the “C1” shaft (the shaft on the far right). Ensure
that it does not interfere with any rotating components. Rest the accelerometer
cord in one of the notches provided and close the safety cover.

4. Turn on the Acoem Falcon device and press “Collect” on the home screen.

5. Select “Fault Analysis Test” and press “Reset.” Press “Yes” when prompted to
confirm the reset. This will delete and previously recorded data.

6. On the Routes list, press “OK.” The device should be on the first step,
“C1 - Baseline 35 Hz.” If not, select “Explore,” then select that step and hit “OK.”

7. Supply power to the motor by selecting “C1 On” with the arrow buttons on the
control panel and pressing the green “ENTER” button. You should hear a loud
click as the switch closes.

53
8. Start the motor by turning on the VFD (green button). With the VFD on, you can
change the speed using the arrow keys. Set the VFD speed to 35 Hz. Wait a few
seconds for the motor speed to stabilize.

Note: The VFD determines the frequency of the signal powering the motor. The
actual motor speed may be slightly lower than this due to motor slip. You will
measure the actual motor speed later on via the software.

9. On the Falcon, press “Acquisition.” It should collect accelerometer data for

approximately 26 seconds. When it finishes, you should see check marks next to
the collected data.

10. Press the red button on the VFD. Wait for the motor to stop. Then cut the power
by selecting “C1 Off” on the control panel and pressing “ENTER.”

11. Add one bolt to the flywheel on the C1 shaft, then close the safety cover again.

12. Re-power the motor (“C1 On”) and turn on the VFD. Wait for the motor speed to
stabilize at 35 Hz.

13. On the Falcon, use the “Next” or “Previous” buttons to select “C1 - 1 Bolt 35 Hz.”
Select “Acquisition” to acquire data.

14. After the data acquisition is complete, stop the motor using the VFD and then cut
power (“C1 Off”).

15. Add another bolt to the flywheel, in the hole opposite the first one, to rebalance
it. Close the safety cover.

16. Turn the motor power back on (“C1 On”) and turn on the VFD.

17. On the Falcon, navigate to step “C1 - 2 Bolts 35 Hz” and select “Acquisition.”

18. After the data acquisition is complete, stop the motor using the VFD and then cut
power (“C1 Off”). Remove both bolts from the flywheel.

Shaft C2

19. Move the accelerometer to the top of the outboard bearing housing of the “C2”
shaft (the middle shaft). Ensure that the accelerometer cable does not interfere
with any rotating equipment and passes through one of the notches, then close the
safety cover.

20. Supply power to the second motor (“C2 On”), then turn on the VFD. Set the motor
speed to 50 Hz and wait a few seconds for the motor speed to stabilize.

21. On the Falcon, navigate to step “C2 - 50 Hz” and select “Acquisition.”

54
22. Once data acquisition is complete, decrease the motor speed to 20 Hz, then turn
off the VFD. Cut power to the motor (“C2 Off”).

Shaft C3

23. Move the accelerometer to the outboard bearing housing of the “C3” shaft where
the missing bolt would be. Ensure that the cable does not interfere with any moving
equipment and passes through one of the notches.

24. Add one bolt to the flywheel on the “C3” shaft to produce a predictable cyclical
load. Close the safety cover.

25. Supply power to the third motor (“C3 On”), then turn on the VFD. Ensure that the
motor speed is still set at 20 Hz and wait for it to stabilize.

26. On the Falcon, navigate to step “C3 - 20 Hz” and select “Acquisition.”

27. Once data acquisition is complete, turn off the motor using the VFD, then cut power
to the motor (“C3 Off”). Remove the bolt from the flywheel. Ask your TA for a
secure location to store the accelerometer.

Data Collection

28. Insert the USB drive into the bottom of the Falcon.

29. On the Falcon, return to the “Routes list” page. With the “Fault Analysis Test” route
highlighted, select “Export.” This will export the collected data as a “.zdl” data file
to the “Export” folder on the USB. Turn off the Falcon device when complete.

30. Remove the USB drive from the Falcon and insert it into the lab computer. Open
the “Export” folder on the USB and identify your data file. Rename the file to
include your group number and today’s date (for example, “S1G3_08292022.zdl”).

31. Open the NEST software on the computer. From the home screen, click “Collect.”
Select “Via intermediate files,” then click on the registered Falcon device.

32. Move to the tab labelled “Instrument -> NEST,” click “Add,” and select your “.zdl”
file from the USB.

33. Click the check box next to the uploaded test on NEST, then click “Send.” Select
“OK” in response to the text boxes.

34. Return to the NEST home screen by clicking the house icon at the top left of the
screen.

35. Select “Analyst.” Ensure you are on the “Vibration Lab - Fault Analysis” asset on
the left. You should see a list of the five data sets you collected. Double-click on
any one of these tests to view the results.

55
 Note: The NEST software stores previous test data so that hardware performance
can be tracked over time. This means you are likely to see other groups’ results
listed. Look at the list of results with their measurement dates and ensure you are
viewing your own group’s data.

36. For each of the five tests, do the following.

a. Double-click the icon next to the acceleration spectrum. This will open a
window containing the acceleration spectrum graph. Maximize the window
and take a screenshot. (If you press [Alt]+[PrtSc], it will take a screenshot
of the current window only and store it on the clipboard. You can then paste
this into Paint or WordPad using [Ctrl]+[V].)

b. Click the “V” button at the top of the spectrum window. This will display the
spectrum in terms of velocity. Take a screenshot of this graph.

c. Click the “D” button at the top of the spectrum window. This will display the
spectrum in terms of displacement. Take a screenshot of this graph.

d. Return to the velocity spectrum graph (“V” button). Identify the largest
peak that is close to the expected motor speed and place a cursor at that
peak. You just need to click near the peak, and the software will identify it
and place the cursor at the appropriate location. This is your 1×TS peak.
Using the ← and → arrow keys on the keyboard, you can manually move
the cursor and identify the resolution of the frequency in the FFT.

e. Record the data reported from the cursor: frequency, acceleration, velocity,
and displacement magnitudes. Take this frequency as the turning speed
(angular velocity) of the motor. (It may be slightly slower than the frequency
you set on the VFD due to motor slip.)

f. Identify any other significant peaks that occur at harmonics of the first
peak (maximum three additional peaks – choose the largest ones). Place
a cursor at each of these peaks and record the data (frequency,
acceleration, velocity, and displacement magnitudes). You may ignore any
peaks with a velocity magnitude less than 25 % of the 1× peak. You may
treat any peaks at frequencies below ~8 Hz as noise.

g. Close the spectrum graph. Double-click on the icon next to the time
waveform. This will open a window containing the recorded acceleration
versus time data from the accelerometer. Maximize the window and take a
screenshot. You may now close the graph.

37. Confirm that you have all of the screenshots (15 spectrum graphs and 5 time
waveform graphs) and have recorded the data from all the cursors. Save your
screenshots to at least two secure locations. Close the NEST software.
Ensure that all three motor connection circuits (C1, C2, and C3) are turned Off.
Shut down the computer.
56
Post-Lab Analysis

38. Compare the vibration velocity frequency spectra from Step 9 (C1 - Baseline),
Step 13 (C1 - 1 Bolt), and Step 17 (C1 - 2 Bolts). Comment on and quantify
and similarities and differences. Can you explain why the differences are
occurring?

39. Compare the vibration velocity frequency spectra from Step 13 (C1 - 1 Bolt),
Step 21 (C2), and Step 26 (C3). For each one, try to identify the fault type from
one of the fault types listed in the Introduction based on the spectrum
characteristics. Justify your answers using quantitative evidence.

40. Discuss your results from Step 39. Do your answers make sense in terms of
the procedure and your observations during the lab?

41. For the “C1 - 1 Bolt” case, calculate the centrifugal force the bolt produced on
the shaft.

42. Assume the force you calculated in Step 41 is the dominant radial force on the
bearing. Calculate the expected bearing life, using a reliability factor of
𝑟 = 95 %. You may assume uniform load, no impact conditions. Report the
bearing life both in number of cycles and in terms of continuous run time.

43. Repeat the calculations in Steps 41 and 42 for the following hypothetical
scenarios. Present the results and briefly comment on them.

a. The mass of the bolt is increased by a factor of 10.

b. The angular velocity of the shaft is increased by a factor of 10.

44. Using your data from the “C3” case, identify the largest acceleration magnitude.
Assuming this acceleration is felt by the outboard bearing and accelerometer
(combined mass 0.642 kg), calculate the corresponding impact force on the
bearing.

45. Assume the force you calculated in Step 44 is the dominant radial force on the
bearing. Calculate the expected bearing life, using a reliability factor of
𝑟 = 95 %. Assume light impact. Comment on your results.

References
[1] S.S. Rao, Mechanical Vibrations, 6th Edition, Global Edition. Pearson, 2011,
pp.942-944.

[2] R.S. Figliola, D.E. Beasley, Theory and Design for Mechanical Measurements, 6th
Edition. Wiley, 2015, pp. 71-72.

[3] R.C. Juvinall, Fundamentals of Machine Component Design, 5th Edition. Austin,
TX, USA: Wiley, 2012.

57
[This page intentionally left blank.]

58
 (G). RESONANCE EXPERIMENT

Introduction
Systems with inertia require a second-derivative term in the equations that
characterize their behavior, meaning they can be classified as second-order (or higher-
order) systems [1]. For example, the behavior of a mass-spring-damper system with one-
dimensional motion can be predicted using Equation 34.
⬚𝑚𝑦̈ + 𝑐𝑦̇ + 𝑘𝑦 = 𝐹 (𝑡)⬚ (34)
𝑒𝑥𝑡
⬚ ⬚
• 𝑚 = Mass of the moving object
• 𝑐 = Damping coefficient
• 𝑘 = Spring constant or stiffness
• 𝑦 = Displacement of the moving object
• 𝐹𝑒𝑥𝑡 (𝑡) = External force on the object (as a function of time)
A second-order system can be characterized by a natural frequency 𝜔𝑛 and a
dimensionless damping ratio 𝜁. For a mass-spring-damper system, the natural frequency
(in rad/s) and the damping ratio can be found using Equations 35 and 36, respectively.

⬚𝜔 = √ 𝑘 ⬚ (35)
⬚ 𝑛 𝑚⬚
⬚ 𝑐 ⬚
⬚𝜁 = ⬚ (36)
2𝑚𝜔𝑛
⬚ ⬚
For static loading, the static displacement 𝛿𝑠𝑡 depends only the external force 𝐹0
and the stiffness, as all time derivatives are equal to zero. The result is Hooke’s law (Eq.
37). However, when the external force is cyclical, the amplitude of the displacement also
depends on the frequency 𝜔 of the force and on the natural frequency and damping ratio
of the system. Furthermore, the displacement will not be in phase with the external force,
but will lag behind by some phase angle 𝜙. The external force and corresponding
displacement are described by Equations 38 and 39.
⬚𝛿 = |𝐹0 | ⬚ (37)
⬚ 𝑠𝑡 𝑘 ⬚
⬚𝐹 (𝑡) = 𝐹 cos(𝜔𝑡) ⬚ (38)
⬚ 𝑒𝑥𝑡 0
⬚
⬚𝑦(𝑡) = 𝑋 cos(𝜔𝑡 − 𝜙) ⬚ (39)
⬚ ⬚
The phase angle can be found using Equation 40. To find 𝑋, the amplitude of the
cyclical displacement, we first find the magnitude ratio 𝑀 using Equation 41. This is a
ratio of the amplitude of the cyclical displacement (𝑋) to the static displacement (𝛿𝑠𝑡 ) that
we would expect for a static force of the same amplitude (𝐹0 ).
⬚𝜙 = tan−1 ( 2𝜁𝑟 ) ⬚ (40)
⬚ 1 − 𝑟2 ⬚
⬚𝑀 = 𝑋 = 1
, 𝑟=
𝜔 ⬚
(41)
⬚ 𝛿𝑠𝑡 √(1 − 𝑟 2 )2 + (2𝜁𝑟)2 𝜔𝑛 ⬚
As the equations show, the magnitude ratio and phase angle depend only on the
damping ratio 𝜁 and on the frequency ratio 𝑟 = 𝜔/𝜔𝑛 , or the frequency of the external
force divided by the natural frequency of the system. For convenience, 𝑀 and 𝜙 are often
59
plotted for reference in what are sometimes called Bode plots (after Hendrik Wade Bode).
Figure 26 shows these plots for a second-order system.

Figure 26. Bode plots: magnitude ratio and phase angle for second-order systems

Resonance is a phenomenon that can occur in second-order systems (or higher-

order systems). A system experiences resonance whenever the magnitude ratio 𝑀 is
greater than one. This means the system is exhibiting more displacement than the
stiffness alone predicts. For second-order systems, resonance can only occur for
damping ratios 0 < 𝜁 < 1/√2 (0 < 𝜁 ≲ 0.707), and only at frequencies in the vicinity of the
resonant frequency, 𝜔𝑅 (sometimes called critical frequency) of the system.
𝜔𝑅 = 𝜔𝑛 √1 − 2𝜁 2 (42)
If the damping ratio is low enough, then excitation near the resonant frequency can
produce vibrations much higher than the stiffness of the system predicts (𝑀 ≫ 1, or
𝑋 ≫ 𝛿𝑠𝑡 ). This can lead to issues in machine or structural design. For example, vibrations
from an idling engine or ambient activity in a building may cause unacceptably large
vibrations in any system (such as a steering wheel or equipment table) whose resonant
frequency matches those vibrations [1]. Resonance in helical valve springs can cause
spring surge violent enough to fracture the springs [2].
Engineers have to predict such situations during the design process. They may
avoid such issues by adding additional damping or by changing the natural frequency of
the system so that the expected external forces do not approach the critical frequency.
The discussion so far has been for second-order systems with a single degree of
freedom, or motion along one direction only. In practice, most systems have more than
one degree of freedom. For example, if a rotating shaft is supported by two ball bearings,
then both bearings act as stiff springs to maintain the position of the shaft centerline. The
system could be modelled as two mass-spring-damper systems (one per bearing), or it
could be modelled as the shaft experiencing two degrees of freedom: radial motion
(moving perpendicular to the axis) plus angular motion (the axis itself becoming no longer
parallel to its original position). When investigating resonant frequencies in a structure
such as a table, the table can have multiple modes of vibration: the tabletop bending, the
60
legs compressing, the table swaying side to side, etc. When these degrees of freedom
are coupled to each other, their analysis becomes complicated. It can still be done, for
example, by drawing free-body diagrams, writing the characteristic equations in matrix
form, and solving eigenvalue problems. However, this is beyond the scope of this lab.
Further reading is available in [3] for those who are interested. Fortunately, in many
instances, the coupling is weak enough that treating each degree of freedom as an
independent second-order system can produce acceptably accurate predictions. In any
case, a system with multiple degrees of freedom may be thought of as having multiple
natural frequencies, any number of which could potentially incite resonance.
In this lab, you will be investigating the vibration analysis system and the lab table
it is sitting on, and their dynamic response to a known cyclical force. Figures 27 and 28
demonstrate the general setup. You will be modelling the table as a rectangular beam
with simply-supported ends and a force 𝐹 at a distance 𝑎 from one end. Using an
accelerometer, you will measure the motion of the system at this location. The effective
stiffness at this location is given by Equation 43, where 𝐸 is Young’s modulus for the table
and 𝑙, 𝑤, and ℎ are the length, width, and height of the table, respectively.

Figure 27. Diagram of mass Figure 28. Simply-supported beam with off-
(vibration analysis system) on table center load

⬚𝑘 = − 𝐹 = 3𝐸𝐼𝑙 ⬚
(43)
⬚ 𝑦 𝑎 (𝑙 − 𝑎)2 ⬚
2
3
⬚𝐼 = 𝑤ℎ ⬚ (44)
⬚ 12 ⬚
The force will be generated by adding a bolt of mass 𝑚𝑏 at radius 𝑟𝑏 to a flywheel
on one of the shafts in the vibration analysis system. When the shaft spins at speed 𝜔,
this will generate a centrifugal force of magnitude:
⬚𝐹 = 𝑚 𝑟 𝜔 2 ⬚ (45)
⬚ 0 𝑏 𝑏
⬚
You will measure the amplitude of the vibrations for a range of shaft speeds, then
plot the magnitude ratio as a function of frequency, and compare against the theoretical
curve.

Objective
In this experiment, students will…
• Calculated centrifugal forces on an unbalanced shaft;
• Record accelerator data measured over a range of rotational speeds;
• Calculate experimental and theoretical magnitude ratios and compare the two.

61
Necessary Equipment and Materials
• Vibration Analysis Modular System (mass = 76 kg), includes:
o Variable-frequency drive (VFD) controller
o Three AC motors on independent contactor circuits C1, C2, and C3
o Siemens programmable panel to open/close circuits
o Three stainless steel shafts, each coupled to a motor (8 inch distance
between ball bearings)
o Two 1 kg brass flywheels (diameter 5 inches, bolt holes at a 1.5 inch radius)
• Bolts for unbalancing flywheels (mass of 24.60 g ± 0.01 g)
• Acoem FALCON handheld vibration analyzer
• Integrated Electronics Piezo-Electric (IEPE) accelerometer
• Laboratory table
o Width: 0.6604 m
o Length: 1.3716 m
o Thickness: 0.0254 m
o Distance from edge of table to C1 shaft: 0.6318 m
o Modulus of elasticity: 44.1 GPa
• PC with NEST i4.0 software installed
• Flash drive for transferring data files

Figure 29. Vibration Analysis Modular System showing locations of shafts and bearings.

NOTE: Keep lab table clear!

During this experiment, you will be measuring vibrations on the Vibration Analysis
System. The lab table is a part of this system and can influence the results at certain
frequencies. Do not lean on or bump the table, nor leave heavy objects on the table,
while you are collecting data, as these may affect your results!
d
62
Procedure
1. Insert the flash drive into the PC. On the flash drive, create a folder named “Export”
if it does not already exist. This will be used to export data files to the NEST
software. If the folder already exists, you may delete any files in the folder.

2. Ensure that none of the motor power circuits are on by doing the following: On the
Siemens control panel, use the arrow buttons to navigate the cursor to “C1 Off,”
then press the green “ENTER” button. Repeat for “C2 Off” and “C3 Off.”

3. Calculate the natural frequency of the table/vibration analysis system in bending

mode using Equations 43 and 35. All necessary dimensions and material
properties are given in the list of equipment.

4. Insert one bolt into the brass flywheel on the “C1” shaft (the shaft on the far right).

5. Place the accelerometer vertically on the outboard bearing housing of the “C1”
shaft. Ensure that it does not interfere with any rotating components. Rest the
accelerometer cord in one of the notches provided and close the safety cover.

6. Turn on the Acoem Falcon device and press “Collect” on the home screen.

7. Select “Resonance Test C1” and press “Reset.” Press “Yes” when prompted to
confirm the reset. This will delete any previously recorded data.

8. On the Routes list, press “OK.” The device should be on the first step, “Resonance
10 Hz.” If not, select “Explore,” then select that step and hit “OK.”

9. Supply power to the motor by selecting “C1 On” with the arrow buttons on the
control panel and pressing the green “ENTER” button. You should hear a loud
click as the switch closes.

10. Start the motor by turning on the VFD (green button). With the VFD on, you can
change the speed using the arrow keys. Set the VFD speed to 10 Hz. Wait a few
seconds for the motor speed to stabilize.

Note: The VFD determines the frequency of the signal powering the motor. The
actual motor speed may be slightly lower than this due to motor slip. You will
measure the actual motor speed later on via the software.

11. On the Falcon, press “Acquisition.” It should collect accelerometer data for
approximately 26 seconds. When it finishes, you should see check marks next to
the collected data.

12. Once the data acquisition finishes, use the “Next” button to go to the next test,
which should be “Resonance 12 Hz.”

13. Set the VFD speed to 12 Hz. Wait a few seconds for the motor speed to stabilize,
then press “Acquisition” on the Falcon to collected the data.

63
14. Repeat Steps 12 and 13 for all the tests indicated by the Falcon. There should be
a total of 14 speeds: 10, 12, 14, 15, 16, 17, 18, 20, 25, 30, 35, 40, 45, and 50 Hz.

15. Once all tests have been completed, you may turn off the motor. Press the red
button on the VFD and wait for the motor to stop. Then cut the power by selecting
“C1 Off” on the control panel and pressing “ENTER.”

16. Remove the bolt from the flywheel. Ask your TA for a secure location to store
the accelerometer.

Data Collection

17. Insert the USB drive into the bottom of the Falcon.

18. On the Falcon, return to the “Routes list” page. With the “Resonance Test C1”
route highlighted, select “Export.” This will export the collected data as a “.zdl”
data file to the “Export” folder on the USB. Turn off the Falcon device when
complete.

19. Remove the USB drive from the Falcon and insert it into the lab computer. Open
the “Export” folder on the USB and identify your data file. Rename the file to
include your group number and today’s date (for example, “S1G3_08292022.zdl”).

20. Open the NEST software on the computer. From the home screen, click “Collect.”
Select “Via intermediate files,” then click on the registered Falcon device.

21. Move to the tab labelled “Instrument -> NEST,” click “Add,” and select your “.zdl”
file from the USB.

22. Click the check box next to the uploaded test on NEST, then click “Send.” Select
“OK” in response to the text boxes.

23. Return to the NEST home screen by clicking the house icon at the top left of the
screen.

24. Select “Analyst.” Ensure you are on the “Vibration Lab - Resonance” asset on the
left. You should see a list of the 14 data sets you collected. Double-click on any
one of these tests to view the results.

Note: The NEST software stores previous test data so that hardware performance
can be tracked over time. This means you are likely to see other groups’ results
listed. Look at the list of results with their measurement dates and ensure you are
viewing your own group’s data.

25. For each of the 14 tests, do the following.

a. Double-click the icon next to the acceleration spectrum. This will open a
window containing the acceleration spectrum graph. Maximize the window.

64
 b. Identify the largest peak on the graph that is at approximately the expected
motor speed (labelled “f0”) and place a cursor at that peak. You just need to
click near the peak, and the software will identify it and place the cursor at the
appropriate location. Using the ← and → arrow keys on the keyboard, you can
manually move the cursor and identify the resolution of the frequency in the
FFT.

c. Record the data reported from the cursor: frequency, acceleration, velocity,
and displacement magnitudes. Take this frequency as the actual turning speed
(angular velocity) of the motor. (It may be slightly slower than the frequency
you set on the VFD due to motor slip.)

26. Identify the test run whose actual speed (the one you measured using the cursor)
most closely matches the natural frequency you calculated in Step 3.

a. Open the displacement spectrum for this speed (double-click the icon) and
maximize the window. Place a cursor at the largest peak near the motor speed,
and verify that the numbers (frequency, acceleration, velocity, displacement)
match what you recorded earlier.

b. Take a screenshot of this window. (If you press [Alt]+[PrtSc], it will take a
screenshot of the current window only and store it on the clipboard. You can
then paste this into Paint or WordPad using [Ctrl]+[V].)

c. Close the spectrum graph. Double-click on the icon next to the time signal.
This will open a window containing the recorded acceleration versus time data
from the accelerometer. Maximize the window and take a screenshot. You
may now close the graph.

27. Confirm that you have the data from all 14 tests plus the two screenshots. Save
your screenshots to at least two secure locations. Close the NEST software.
Ensure that all three motor connection circuits (C1, C2, and C3) are turned Off.
Shut down the computer.

Post-Lab Analysis

28. Calculate and plot the experimental magnitude ratio versus frequency for the 14
tests. For each test, you will need to calculate the centrifugal force 𝐹0 from the
rotating bolt and the corresponding expected static deflection 𝛿𝑠𝑡 . The amplitude
𝑋 of the displacement comes from the data you measured on the graph. Include
a table of these values somewhere in your report (it may be in an appendix).

29. Using the measured data point from Step 26 (the one that was closest to the
theoretical natural frequency) and Equation 41, calculate the estimated damping
ratio and damping coefficient and report both values.

65
30. Using the damping ratio you calculated in Step 29, calculate and plot the
theoretical magnitude ratio for all frequencies from 0 to 50 Hz. Add this to your
plot from Step 28.

31. Comment on the theoretical (Step 30) and experimental (Step 28) magnitude
ratios. Make both quantitative and qualitative comparisons. Are there any
conclusions you can draw from your data?

32. Assuming the displacement of the bearing housing can be described by

Equation 39, derive an expression for the acceleration 𝑎(𝑡) by taking the second
derivative.

33. Using the measured data point from Step 26 (the one that was closest to the
theoretical natural frequency), calculate all the terms in the acceleration expression
you derived in Step 32. Plot this equation for 0.00 𝑠 ≤ 𝑡 ≤ 0.64 𝑠.

34. Compare your plot of semi-theoretical acceleration vs. time (Step 33) to the
screenshot of the measured acceleration vs. time (Step 26.c). Make both
quantitative and qualitative comparisons. How well do they agree with each other?
Are there any conclusions you can draw from your data?

References
[1] R.S. Figliola, D.E. Beasley, Theory and Design for Mechanical Measurements, 6th
Edition. Wiley, 2015.

[2] J.E. Shigley, C.R. Mischke, Mechanical Engineering Design, 5th Edition. McGraw-
Hill, 1989, pp. 433 – 435.

[3] S.S. Rao, Mechanical Vibrations, 6th Edition, Global Edition. Pearson, 2011,
ch. 5 – 6.

66
 (H). FIBER-BRAGG GRATING EXPERIMENT

Introduction
A fiber-Bragg grating is a type of sensor embedded in a fiber-optic cable which can
be used for a wide variety of measurements, including strain and temperature
measurement. The sensor consists of a series of bands in the fiber-optic cable where the
glass has been treated to locally modify the refractive index. This creates a number of
semi-reflective surfaces in the cable. When broad-spectrum light is passed through the
fiber-optic cable, most of the light will eventually be transmitted through. However,
components of the light whose wavelengths are sufficiently near the so-called Bragg
wavelength, 𝜆𝐵 , will be largely reflected back to the source. The Bragg wavelength is,
typically, directly proportional to the spacing between the interfaces, 𝐿.

⬚𝜆 = 𝛼𝐿⬚ (46)
⬚ 𝐵 ⬚

Figure 30. Fiber optic Bragg grating [1]

When the spacing in the grating changes, such as through loading of the fiber
through tension/compression or through thermal expansion/contraction, the Bragg
wavelength will therefore change, as well. This is how the sensors are used. An
interrogator is used to create broad-spectrum light, transmit it through the fiber-optic
cable, and then measure the wavelength spectrum of the light that is transmitted back. A
large peak should occur around the Bragg wavelength, as shown in Figure 1. The
interrogator then quantifies this wavelength, which can be converted to a strain,
temperature, or other quantity, depending on how the sensor is being loaded.
For strain measurements, the conversion is simple. Since the Bragg wavelength
is proportional to 𝐿, which is proportional to the length of a segment of the fiber, the
relative change in wavelength gives the strain. However, the strain experienced by the
fiber may not be the same as that experienced by the part we are trying to measure due
to differences in material properties and/or bonding issues. For this reason, a gauge
factor 𝑓 is introduced, as shown in Equation 48.

67
 ⬚ Δ𝜆𝐵 = Δ𝐿) ⬚ (47)
⬚ 𝜆𝐵 𝐿 𝑠𝑒𝑛𝑠𝑜𝑟 ⬚

⬚𝜀 Δ𝜆𝐵 ⬚
𝑎𝑐𝑡𝑢𝑎𝑙 = (48)
⬚ 𝑓 ⋅ 𝜆𝐵 ⬚

One feature of fiber-optic sensors is the ability to include multiple signals in a single
fiber-optic cable. The only constraint is that the nominal wavelengths of each grating in
a single cable must be sufficiently far apart from each other to ensure there is no overlap
in their signals.
Since FBG sensors are sensitive to strain caused by thermal expansion as well as
external loading, there are instances where it is necessary to measure and account for
this strain. A simple way to do this is to include another grating that is not subjected to
the physical loading. Thus, any changes in the wavelength of this sensor should be due
to thermal effects. By subtracting the effective strain measured by the unloaded sensor,
we can compensate for these temperature changes.
In this lab, you will simulate a wing-loading test on a 3D-printed model wing. The
wing model has a fiber-optic cable mounted to it with five FBG sensors created in it. Four
will be used to measure strain on the wing, and one will be used as temperature
compensation. You will measure both the wavelengths and the reported strain from these
sensors as a function of the load applied to the wing. Tests like this are sometimes used
to monitor the health of structures. If the strain on a structure starts behaving differently
under the same loading conditions, it could be a sign of a defect forming.
You will compare your experimental results against both hand calculations and an
FEA model. The hand calculations are based on simple beam theory, treating the wing
as a cantilever beam. You should be familiar with the equations for strain in a beam
already. For the numeric simulations, you will be using ANSYS to model the stresses
and strains in the wing. A model of the wing has already been prepared for you. You
will, however, need to adjust the applied load, as well as the material properties.
The fact that the wing material was created using additive manufacturing presents
its own set of challenges. 3D-printed materials typically do not demonstrate the same
material properties as their bulk materials, which may be partly due to voids in the printed
part. The material properties are also frequently anisotropic. For example, the ultimate
strength of thin-wall materials when tension is applied 90° to the print layers is often much
lower than when tension is applied along the layers. One study [2] found a particular
batch of 3D-printed polylactic acid (PLA) parts to have a Young’s modulus of
(2970 ±102) MPa and a Poisson’s ratio of 0.333 ±0.002 when tested at 90° to the print
layers.

68
Objective
In this experiment, students will…
• Measure the wavelength and reported strain at four locations on a wing using FBG
sensors as a function of load applied at a given point;
• Verify the strain calculations of the interrogator software using FBG theory;
• Use historical data to compare against their data and draw conclusions about the
structural health of the wing;
• Calculate theoretical strain using basic beam theory and compare against the
measured data;
• Calculate theoretical strain using ANSYS models and compare against the
measured data;
• Discuss the validity of the various models as well as weigh their pros and cons.

Necessary Equipment and Materials

• Wing-loading testing rig, includes:
o Aluminum frame with polycarbonate protective shell
o Wing model made of 3D-printed polylactic acid (PLA), with walls two print-
layers thick (approximately 0.0325”)
o Fiber-optic cable containing five FBG sensors (nominal wavelengths
1556 nm, 1560 nm, 1564 nm, 1568 nm, and 1572 nm)
(Check the equipment or ask your TA for a gauge factor.)
o Digital scale (1 g precision) and lab jack with mount to apply load to the
underside of the wing
• Hyperion si155 4-channel interrogator from Luna, Inc.
o 1460 – 1620 nm measurement range
o Standard 1000 Hz scan
• Laptop computer with ENLIGHT software to communicate with interrogator

NOTE: Sensitive equipment!

The 3D-printed wing, the fiber-optic cables, and the Hyperion interrogator are all
delicate pieces of equipment. Be sure not to touch the wing, the sensors, or the
connecting cables to avoid damaging them. In addition, do not place anything on the
shell of the wing frame nor attempt to move the frame. Pay attention when applying
loads not to exceed the limits specified.

Procedure
1. Verify that the load applicator on the jack is not in contact with the wing. If it is,
notify your TA.

2. Turn on the laptop and the Hyperion interrogator. Wait for the interrogator to
complete its configuration and set an IP address.

3. Launch the ENLIGHT software on the laptop. Open the configuration file
WingLoading_v2.moi. On the Acquisition panel, you should see the signal from
Ch. 1, with five peaks visible near the five nominal wavelengths. Take a
69
 screenshot of this panel. (Note: There is a screenshot tool in the ENLIGHT
software. This saves files to the folder Documents\Micron Optics\ENLIGHT\Data.)

4. Click on the Sensors tab. On the top portion of the panel, you should see the five
currently-measured Bragg wavelengths (along with their detection ranges) and a
visual representation of the five sensor peaks. If any of the sensors are not
showing a wavelength, if there are not five peaks visible on the graph, or if any of
the peaks are below ~-50 dBm, notify your TA before continuing.

5. On the bottom portion of the Sensors panel, you should see the five sensors
(labelled STG1 through STG4 and TEMP5) with their wavelengths converted to
strain values in µε. Click the button on the right of the screen that indicates “Zero
all sensors.” This will set reference values for the sensors. The current strain
measurements should now be close to zero (±10 µε).

6. Take a screenshot to record the instantaneous wavelength and strain values for
all five sensors (10 values total). Be sure to record the exact time given in the
filename so you can identify the correct screenshot.

CAUTION: WHEN LOADING THE WING, DO NOT EXCEED 1200 GRAMS AT

ANY POINT!

7. Have your TA verify that the scale is in the correct position. Slowly turn the knob
to raise the jack until it just comes into contact with the underside of the wing.
Adjust the knob until the force (when you are not touching it) reads 250 g ±50 g.
You may need to wait a few seconds while the force stabilizes.

8. Simultaneously record the weight (in gram-force) shown on the scale and take a
screenshot of the Sensors panel. You will need to record the measured
wavelengths and strain values of the five sensors as before.

9. Slowly turn the knob until the scale reads 500 g ±50 g and let it stabilize.
Simultaneously record the weight and take a screenshot to record the sensor
values.

10. Repeat Step 9 for 750 g ±50 g, and then for 1000 g ±50 g (remembering not to
exceed 1200 g).

11. Slowly turn the knob in the other direction until the jack is in the fully lowered
position and is not in contact with the wing. Take a final screenshot to record the
wavelengths and strain values of the five sensors. Turn off the digital scale.

12. After verifying you have all of the data, close the ENLIGHT software (do not save
the configuration file), then turn off the Hyperion interrogator.

13. Save your screenshots to a secure location, then delete them from the laptop.

14. Download a copy of the zipped ANSYS project files from the laptop. You will use
these after the lab.
70
15. Turn off the Hyperion interrogator and the laptop.

Post-Lab Analysis

16. For each of the sensors at the ~1000 g loading case, calculate the experimental
strain using the measured wavelengths and Equation 48 (where Δ𝜆𝐵 is the change
in wavelength relative to the no-load case). Compare these values against the
strain measurements given by the software.

17. For sensors STG1 to STG4, plot the measured strain versus load on a single
graph. Apply a proportional fit (𝑦 = 𝑚𝑥) to each. Discuss the validity of assuming
a proportional relationship between load and strain for the test data. Use
regression statistics to back up your claims.

18. Download the reference data posted to Moodle. Compare the unloaded
wavelengths to the unloaded wavelengths you recorded, and discuss. Do the
following:

a. For each strain gauge for each of the loaded tests (should all be ~1000 g
load), calculate an effective local bending compliance by dividing the
measured strain (in µε) by the applied load (in gf).

b. Calculate a 95 % precision interval of effective compliance values based on

the loaded reference tests.

c. For your ~1000 g loading case, calculate the effective compliance for each
of the strain gauges (STG1 to STG4). Present your values with the
precision intervals and discuss.

19. For the ~1000 g loading case, perform hand calculations to predict the strains at
STG1 (Y-direction) and STG2 – 4 (Z-direction). The location of the load, locations
of the sensors, and area moments of inertia at the sensor locations are given on
the following pages.

20. On a computer with ANSYS installed (such as the CADLab computers), unzip the
WingModel.zip file. Launch ANSYS and open the file “Wing 2023-03-26.wbpj”.

21. Double-click on the Engineering Data step (#2) to open the materials list. Click on
PLA, then expand the Isotropic Elasticity item by clicking the + sign. Click on
Young’s modulus. Double-click the value in the table, then change the number to
the lower limit of the expected Young’s modulus based on the values given in the
Introduction. Press Ctrl-S to save, then close the Engineering Data tab.

22. Double-click on the Setup step (#5) to open the ANSYS Mechanical tool. If
prompted to re-read upstream data, select OK. It may take a moment to load.
Under Static Structural (A5), click on the Force object. In the table that appears
below, enter the magnitude of the applied force for your ~1000 g case. (Note, the
force should be negative and in the X-direction.) Pay attention to the units.

71
23. Under Model (A4), click the + sign next to Geometry to expand it, then expand
SYS. There should be 28 objects shown under SYS. Select all 28 objects using
the mouse and Shift key. In the table that appears below, click on the Thickness
value and change it to 0.0325 in. (or 0.8255 mm), then press Enter. This
represents the thickness of the printed surfaces.

24. Click the “Solve” icon (with the lightning bolt) at the top of the window. If you do
not see it, go to the Environment tab. ANSYS should begin solving the problem.
This may take a few minutes.

25. Once the solution is ready, click on Strain Probe under Solution (A6). In the table
that appears below, read and record the normal strain in the Y direction.

26. Click on Strain Probe 2, Strain Probe 3, and Strain Probe 4, record the normal
strains in the Z direction.

27. Click on Equivalent Stress. This will show you the stress field on the wing. It is
likely that the maximum stress will be near where the force was applied. Adjust
the scale of the contour bar until the stresses on the top of the wing become easy
to see. You may adjust the camera view as desired. Take a screenshot to include
in your report.

28. Close the Mechanical window. Repeat Step 20, but change the Young’ modulus
to the upper limit of the expected values. Repeat steps 21 – 25 to get the new
predictions.

29. Compare the results of your measured strain to those you predicted using hand
calculations and those you predicted using ANSYS. Discuss the differences,
including possible sources of error and which results you trust more.

References
[1] R.S. Figliola, D.E. Beasley, Theory and Design for Mechanical Measurements, 6th
Edition. Wiley, 2015.

[2] X. Wang, L. Zhao, J. Y. H. Fuh and H. P. Lee, "Effect of Porosity on Mechanical

Properties of 3D Printer Polymers: Experiments and Micromechanical Modeling
Based on X-Ray Computed Tomography Analysis," Polymers, vol. 11, no. 1154,
2019.

72
 Theoretical Model Data

Use the following data to calculate the theoretical strains at the strain locations
using simple beam theory.

Bending Height from

Dihedral IY at sensor
Sensor moment arm neutral axis
angle 𝝓 (°) plane (in4)
(in.) (in.)
STG2 4.4 15.97 0.559 0.0982
STG3 4.4 11.12 0.516 0.0855
STG4 4.4 6.24 0.4414 0.0621

Assume that the bending moment arm is at the dihedral angle of the wing, so that the
bending moments are given by:

⬚𝑀 = 𝐹𝐿 cos 𝜙 ⬚ (49)
𝑏
⬚ ⬚

Figure 31. Dihedral angle 𝝓 between neutral axis of wing and horizontal plane

73