Mechanical Engineering

Laboratory Manual
GENERAL
The laboratory section of this course is designed to illustrate experimental procedures in
Mechanical Engineering using simple laboratory devices and instrumentation. For the most part
the experiments will illustrate concepts developed in the lecture course and its prerequisites and
will, therefore, require little or no theoretical introduction. However, some research may be
required to fully understand your experiments. Experiments will be chosen by the course
instructor to be run in laboratory sessions. The student is to prepare for these experiments by
studying the descriptions contained in the laboratory manual or handout prior to attending the
"lab". Each student is expected to participate in each experiment as a member of a "lab group"
as assigned by the instructor.
You will have a chance to manipulate the equipment to see how varying the different
parameters will affect the results of an experiment. You will also have the opportunity to
analyze the processes that take place in an experiment and to compare your theoretical results
with those obtained experimentally. We hope that your powers of analysis will be sharpened by
your laboratory experiences and that the principles presented in class will take on a new
meaning. You will also gain practice in writing effective reports.
Each experiment is to be reported in an engineering report or engineering memorandum
format, which are described later. Your instructor will tell you which to use. You will be graded
on:
Your ability to report your results clearly and concisely (in good English)
Your analysis and interpretation of your results
The conclusions reached
Your recommendations for improving the experiment
Your adherence to the specified format for the report or memorandum
An analysis of the accuracy of your results and a discussion of possible errors
Your course instructor or laboratory instructor will explain the amount each section weights
toward your final report grade. Failure to follow instructions will necessitate a resubmittal of the
documentation. In such cases the instructor will assess a penalty. The lab report due date will be
as shown on the lab schedule or as stated by your instructor. A penalty, to be determined by the
course instructor, will be assessed for late submittals.
Final grades will not be submitted for any student who has not performed all assigned
experiments and submitted all assigned reports. This applies whether or not absences are
excused. After missing an experiment for any reason, you should report to your instructor
as soon as possible (not waiting until the next regular meeting of the class) and arrange to
make up the work with another laboratory section if possible.
LABORATORY CLASS PROCEDURE
In order that the laboratory session is conducted in the most meaningful manner possible,
it is imperative that each student read, study, and understand the experiment to be conducted
prior to coming to the class. The student should also read and understand the laboratory safety
guidelines for the ME-undergraduate laboratories. Failure to follow safety guidelines will result
in expulsion from the lab.
It is necessary that each student brings to class the manual (handout), references if
needed, specified safety equipment not provided in the lab, calculator, graph paper, triangles, etc.
All of these may be necessary to complete your work.
The normal lab session will consist of four parts. These are summarized as follows:
1. A brief lecture by the lab instructor, which will explain the experimental procedure,
safety issues and general theory.
2. A brief Pre-Lab exam covering the background material and lab handout.
3. The conducting of the experiment by groups to collect the data.
4. Calculations and preparation of data for the laboratory report. This is very important
because it will allow identification of erroneous data in a timely manner so that the
experiment may be repeated if necessary.
NOTE: Laboratory data sheets should be signed and dated by the lab instructor before
the student leaves the lab area.
LABORATORY ORGANIZATION
For the laboratory your class may be divided into smaller groups. You will work with the
same group each lab session. The use of smaller groups is to allow each member of the
laboratory group to become more involved in the experiment and to obtain a more
comprehensive understanding of the experiment. The group will normally consist of from five to
seven persons depending on the size of the class. Groups will remain intact for the duration of
the course and will function as a unit in the performance of all experiments. Each member of the
group will be assigned each week to one of the following positions:
Recorder
Operator
Computer
Helper
Assignments will be rotated each week so that each student will serve in each capacity
during the course if possible. The assignment of experiments and of group positions will be
made early in the course, and each student is expected to come to each laboratory period with
definite plans for the work to be done.
Duties of Recorder
The recorder is the leader of the group, and is responsible for the performance of the
assigned experiment in an orderly manner. While each member of the group will be held
responsible for proper preparation of the assignment and for the performance of his duties in the
laboratory, the responsibility for the coordination, general execution of the experiment, and
compliance with all safety requirements rest upon the recorder. He should, therefore, consult the
instructor immediately if delays occur or if there is doubt regarding the method of procedure or if
the computer finds that some data are in error.
The recorder will enter all data on a data sheet, making a copy for the instructor. Before
any test date are taken, he should record the incidental information, such as:
Group personnel present and duties
Data
Title of experiment
Itemized list of the equipment
The recorder should also secure the instructor's approval of the laboratory set-up and
required data before testing is begun. Only original observations are to be recorded as
observed data. No erasures should be made in recording data: instead, a single line should be
drawn through incorrect readings. Calculations made during the progress of the test will be
entered in separate columns and clearly labeled as results. When all the observed data have been
recorded, the recorder will submit the data sheet to the instructor for approval.
Duties of Operator
The operator is responsible for the care and proper functioning of all equipment except
the measuring devices used by the observer. Before any data are taken, he should familiarize
himself thoroughly, under the direction of the instructor, with the operation of the equipment in
order to avoid possible difficulties during the test. He should report any broken or damaged
equipment to the instructor immediately. He should clean up around the equipment when the
testing is completed and see that all equipment is left in the proper condition. The
laboratory equipment should be left clean, orderly, and in suitable condition for another group to
run an experiment.
Duties of Observer
The observer will be responsible for the measuring apparatus. He will take the necessary
readings and will be held accountable for their accuracy. After testing is completed, he will wipe
clean all instruments and leave them in good order for the next group.
Duties of Computer
The computer is responsible for whatever preliminary calculations may be necessary. He
will determine the probable range of values to be measured and will select suitable increments or
observational intervals. These are to be checked with the instructor before the test is started.
During the progress of the experiment, the computer will assist in taking readings, if necessary,
and will cooperate with the recorder in the calculation of results.
Duties of Helpers
The helpers will assist in the experiment by performing duties as designated by the
recorder, such as timer, helping with computations, operation, cleaning, safety procedures or
measurements.
REPORTS
Each student will submit a complete written report covering each experiment performed.
The report is to be the student's own work. The format of the report is explained under the
sections on the Formal Report and the Engineering Memorandum.
The emphasis in the preparation of reports is placed on the accuracy of results. In order
to complete the report in a minimum of time that the experimental data taken in the laboratory
should be reduced and the calculations made prior to leaving the lab. This allows time to
ascertain the accuracy of the data and completeness of the experiment and to rerun the
experiment if necessary.
The report will be written in the third person, past tense (for procedures executed, data
taken, and results obtained), and should be self-sufficient. In other words, the reader should not
need to consult the references in order to understand the report. Correct English and spelling
should be used. The reports are practice for writing technical reports similar to those, which are
required by engineers engaged in industry and engineering practice.
The report must be typed using a word processor. All pages, equations, figures, graphs,
and tables must be numbered. Figures, tables, graphs, etc., must have titles and be introduces in
a sentence in the text of the report. Figures must have axis labels that name the variable as well
as giving its symbol and units if appropriate.
Figures, graphs, and tables must be neat and clear. Figures and graphs should be
generated on the computer through drawing and plotting software (SigmaPlot or Excel are
examples). Choose scales that are appropriate to the range of data and that can be easily read.
Leave room on the paper for scales, labels, and titles. Remember that anything printed within
one and one-fourth inches of the left edge of the paper is almost impossible to read when the
report is bound. Sample figures and graphs are contained in this document.
Instruction for Formal Lab Reports
I. General Comments
Usually engineering tests are summarized in the form of reports. In many cases, these
reports are submitted to supervisors who have not been actively engaged in the test; hence the
reports must be clear and concise enough to leave no doubt concerning the method of test and the
interpretation of the results. It is therefore of the utmost importance that an engineering student
learns to write good reports.
In general, a report is written in the past tense and in the third person. It should be
impersonal throughout, personal pronouns being avoided. The report should be written under the
assumption that none of the readers will be familiar with the work covered. Thus the report must
be completed in itself. The student too frequently is careless in this respect and writes a report
that cannot be followed without a previous knowledge of the test under construction. A good
report is thorough, concise, orderly, neat, and grammatically correct.
II. General Specifications
In order to observe the accepted rules of good writing form, the following specifications
for the general makeup of the report are suggested:
1. Use 8 x 11inch white paper.
2. Write the report with a word processor (available in the SOE computer lab)
3. Use one side of the paper only.
4. Create all drawings and figures using computer drawing and plotting programs. Scanned
images are allowed where appropriate.
5. Use the same font style on drawings and graphs as used in the text. Graphs axes should
be clearly labeled, including units where appropriate.
6. Discrete experimental data that are plotted on appropriate graphs should be designated
with small symbols, such as circles, to distinguish these data from those represented by
curves fitted through them either intuitively or statistically or by mathematical model. If
more that one dependent variable (ordinate) is presented on a graph, each variable should
have a different symbol.
7. When mathematically fitting curves to experimental data, use appropriate judgment. Just
because a 6
th
order polynomial can fit exactly to 7 points does not mean that it is the
appropriate curve for this experimental data (i.e. the distribution may actually be linear or
quadratic). Instead look at the trend of the data and avoid the pitfall of many students in
letting the computer chose the best curve fit. As a general rule, the lower the complexity
of the curve fit that represents the data trends, the better.
III. Report Outline
The material covered and its arrangement in a report varies somewhat with the type of
work being reported. For the usual tests made in a mechanical engineering laboratory, the
following report outline is required for content and order of presentations.
i. Title page
ii. Table of contents
iii. List of Figures
iv. List of Symbols
A. Objective and Introduction
B. Theory
C. Laboratory Procedures
D. Experimental Data
E. Analysis of Results
F. Answers to Questions
G. Conclusions
H. References
I. Appendices (includes original data sheet)
IV. Discussion of Report Outline
A discussion of each section of the report outline follows:
I. Title Page a sample title page attached
II. Table of Contents This should list the remaining parts of the report and give
their page numbers.
III. List of Figures - This should give title and page number for every figure and
graph in the report.
IV. List of Symbols - all symbols used in the report, their definitions, and their units.
Dimensions should be in alphabetical order a through z, followed by Greek
symbols through . Lower case should precede upper case; e.g., if both a and
A. Consistently used subscripts and superscripts may follow the main list of
symbols. For and o is used as a subscript to indicate out, then after the last
symbol show:
Subscripts
Symbol Definition
i in
o out
Then a variable, say T for temperature, may be entered in the symbol list above simply as
Symbol Definition Units
T Temperature C, K
And separated entries are not required for T
i
and T
o
.
A. Objective and Introduction
The objective should be a concise statement setting forth the aim and scope of the investigation.
B. Theory
Background theory for the test should include expected results and derivations of all equations
used.
C. Procedure
In his section a general description of the procedure should be given. This description should be
comprehensive, but brief. It should include a generic list of equipment used and a sketch to
show how the equipment items are related. The enumeration and detailed description of
multitudinous mechanical operations or sequence of such operations such as closing switches,
reading instruments, turning knobs and so forth, should in general be avoided. However, when a
specific method of mechanical operation or sequence of such operations is necessary in order to
insure the validity or accuracy of the test data, it is important that the essential details be included
in the description.
Note that it is unacceptable to simply use or copy the procedural instructions from your lab
handout because:
1. They are written in the second person (the you is implied)
2. They may include many details that are necessary to tell how you perform the experiment
but are unnecessary for the report readers understanding of the procedure.
D. Experimental Data
(a) Data
(1) Name-Plate Data
The first item under data should be the nameplate data or equivalent, of the apparatus
used. The equipment and apparatus should be specified as thoroughly as possible so that
the reader may replicate the experiment or make a value judgement concerning the
accuracy and validity of the experiment based on his knowledge of the equipment.
(2) Observed and Calculated Data
All observed and calculated data should be tabulated when possible. If it its necessary to
correct the raw data to standard conditions, or to make corrections for calibration
errors, the corrected data as well as the raw data should appear in the table. All related
data, whether observed or calculated, should be included in a single table. Headings and
subheadings (titles) identifying individual items of data or sets of data should be used
whether or not the data are tabulated. Instrument identification numbers and ranges need
not be copied from the sheet should be an Appendix of the report.
(b) Sample Calculations
This section should consist of a sample of a complete calculation of each type involved in
the determination of calculated data and the solution of problems. These sample
calculations should first be shown in symbolic form with all symbols properly defined.
Then numerical data should be used, with units shown, in the actual calculations.
Note: When a succession of calculations are required to reach a final result, the same set
of observed data should be used in carrying through the successive sample calculations.
(c) Curves
All curve sheets should conform to the following specifications. A sample curve is
attached.
1. Use a good quality-plotting program. The graph should be plotted on 8 x 11 in. paper
unless otherwise specified.
2. Plot in the first quadrant unless negative values of either the dependent or independent
variable are meaningful.
3. Include scales, axis labels, and figure titles on the graph. MAke sure the graph margin is
1 inch on the top, bottom and right sides of the paper. The left side margin should be 1
inches to allow sufficient space for binding.
4. Plot the independent variable as abscissa (x-coordinate) and the dependent variable as
ordinate (y-coordinate).
5. In general, start the scales of the dependent variable and of the independent variable at
zero.
6. Choose scales that are easy to use and that do not allow points to be plotted to a greater
accuracy than justified by the accuracy of the data.
7. Indicate the points plotted from data by small circles or other symbols.
8. Draw, plot, or calculate a smooth average curve through the plotted points, except in
cases in which discontinuities are known to exist.
Note that the average curve does not necessarily pass through every datum.
Also, if during the performance of the experiment, there was strong evidence that an
equipment malfunction or a procedural error affected a datum, it is appropriate to
disregard that point when drawing the curve. Such a point should be consistently
disregarded in all results affected by it. Statistically only one point is allowed to be
treated in this manner.
9. Place a title containing all pertinent information on each curve sheet.
10. Draw or plot only related curves on the same sheet. Keep in mind that when curves truly
are related it is frequently helpful to interpretation to present them on the same sheet. For
example, for a gasoline engine torque, power, and specific fuel consumption are all
functions of speed; and the engine performance is more easily interpreted if they are all
on the same graph with separate ordinate scales for each variable.
11. Insert curve sheets in the report so that they can be read form the bottom or right side.
Analysis of Results
The analysis of results is the most important section of the entire report. As the
name implies, it should be a complete discussion of the results obtained. Part of this discussion
should deal with the accuracy or reliability of the results. It is suggested that this section consist,
when applicable, of a careful treatment of the effect upon the results of the following:
1. Errors resulting from the necessity of neglecting certain factors because of physical
limitations in the performance of the test
2. Errors in manipulation
3. Errors in observation
4. Errors in instruments
5. Comparison of the results obtained with those that would reasonably have been
expected from a consideration of the theory involved in the problem. Whenever the
theory is apparently contradicted, the probable reasons should be discussed.
When results are given in graphical form as curves the shape of each curve should we
carefully explained. Such an explanation should state the causes for the particular shape the
curve may have. It is not sufficient simply to state that a particular curve has positive slope, the
reason for such a slope should be given. If the slope is not constant, - that is, if the curve is not a
straight-line -its nonlinearity should also be explained. Most of the graphs that you discuss in
this section will have been presented in Experimental Data. Any new graphs resulting from
Analysis of Results should be placed as near the part of the text that discusses them as possible.
Any original conclusions drawn as a consequence of the laboratory procedure and a study
of the results obtained should be given in this section and should be justified by the discussion.
Constructive criticism of any phase of the problem that may seem pertinent may also be
included here.
E. Questions
This in not a normal technical report section, but is included as a place to put the answers
to any specific questions that may be included as part of the experiment. If no specific questions
are asked, do not put a Questions section in your report.
F. Conclusions
In this section the conclusions which were supported and drawn in Analysis of Results
are succinctly restated, usually as a numbered list. No new information should appear in this
section. All justification of conclusions should have occurred in prior sections.
G. References
Publication or other authorities which you use to help you explain the experiment,
calculate results, explain errors, draw conclusions, etc., should be acknowledged using the
format specified by the American Society of Mechanical Engineers (ASME). These
specifications are listed below and can be located on the ASME we site (http://www.asme.org).
If only certain pages of a book, magazine, or paper are actually used, the page range should be
included.
Crediting references must follow two procedures:
A text citation within the body of the report
A Reference page or reference section of the report
Text Citation
Within the text (sections like Introduction, Sample Calculations, and Discussion of Results),
references should be cited by giving the last name of the author(s) and the year of publication of
the reference. The year should always be enclosed in parentheses; whether or not the name of the
author(s) should be enclosed within the parentheses depends on the context. The two possibilities
are illustrated below:
It was shown by Prusa (1983) that the width of the plume decreases under these conditions.
or
It has been shown that the width of the plume decreases under these conditions (Prusa, 1983).
In the case of two authors, the last names of both authors should be included in the citation,
as shown in the above examples, with the word "and'' separating the two authors.
In the case of three or more authors, only the last name of the first author of the reference
should be included, as shown in the above examples, with the other authors being denoted by
"et al.''
In the case of two or more references with the same author(s) and with the same year of
publication, the references should be distinguished in the text by appending a lowercase letter
"a'' to the year of publication of the first cited, a letter "b'' to the second cited, etc. The
references should follow the examples shown above.
List of References or Reference section
References to original sources for cited material should be listed together at the end of the paper;
footnotes should not be used for this purpose. References should be arranged in alphabetical
order according to the last name of the author, or the last name of the first-named author for
papers with more than one author. Each reference should include the last name of each author
followed by his initials.
(1) Reference to journal articles, papers in conference proceedings, or any other collection of
works by numerous authors should include:
Year of publication
Full title of the cited article
Full name of the publication in which it appeared
Volume number (if any)
Inclusive page numbers of the cited article
(2) Reference to textbooks, monographs, theses, and technical reports should include:
Year of publication
Full title of the publication
Publisher
City of publication
Inclusive page numbers of the work being cited
In all cases, titles of books, periodicals, and conference proceedings should be underlined or in
italics. A sample list of references in which these forms are illustrated follows.
Sample References
Kwon, O. K., and Pletcher, R. H., 1981, "Prediction of the Incompressible Flow Over a
Rearward-Facing Step,'' Technical Report HTL-26, CFD-4, Iowa State Univ., Ames, IA.
Lee, Y., Korpela, S. A., and Horne, R. N., 1982, "Structure of Multi-Cellular Natural Convection
in a Tall Vertical Annulus,'' Proceedings, 7th International Heat Transfer Conference, U. Grigul
et al., ed., Hemisphere Publishing Corp., Washington, D.C., Vol. 2, pp. 221-226.
Sparrow, E. M., 1980a, "Fluid-to-Fluid Conjugate Heat Transfer for a Vertical Pipe - Internal
Forced Convection and External Natural Convection,'' ASME Journal of Heat Transfer, Vol.
102, pp. 402-407.
Sparrow, E. M., 1980b, "Forced-Convection Heat Transfer in a Duct Having Spanwise-Periodic
Rectangular Protuberances,'' Numerical Heat Transfer, Vol. 3, pp. 149-167.
Tung, C. Y., 1982, "Evaporative Heat Transfer in the Contact Line of a Mixture,'' Ph.D. Thesis,
Rensselaer Polytechnic Institute, Troy, NY.
I. Appendices
Materials that support the report but are not essential to the readers understanding of it
are included here. The laboratory data sheet should be an appendix. You may wish to use the
laboratory instructions as an appendix. Your complete calculations may be included as an
appendix.
Sample Title Page
Department of Mechanical Engineering
Fluid Mechanics Laboratory
Experiment No.
Internal Flow Meters
ME 312
Lab Group No. ______ Date Performed __________
____________________________ Report Submitted By:
____________________________ ________________________
____________________________ Date Submitted: __________
____________________________ Date Received: ___________
____________________________ Grade: __________________
Sample Graph
This illustrates an acceptable method of showing two curves on one graph. Note that
these curves are closely related in theory. Note that the figure has a title in words. The
axes are plainly marked. The scales chosen are easy to use. Each axis has a label, in
words, and units given. Symbols and line codes are defined. Symbols are used around
discrete data points and the curves pass smoothly near, but not necessarily through, the
data points. To achieve this smoothness a mathematical curve fit procedure was used
(quadratic curve).
Figure 1. Engine Brake Torque and Power as Functions of Speed
0
50
100
150
200
250
300
350
400
450
500
1000 1500 2000 2500 3000 3500
Engine Speed, RPM
B
r
a
k
e

P
o
w
e
r
,

k
W
0
500
1000
1500
2000
2500
B
r
a
k
e

T
o
r
q
u
e
,

N
m
Power
Torque
Torque Trendline
Power Trendline
The Engineering Memorandum
Earlier you were given the format of a formal laboratory report. Many of those
comments apply to the memorandum, which is a shorter, less formal document. A in all
technical writing, attention must be given to the reader. In the case of a memorandum,
assume that you are writing to an engineering supervisor whose technical knowledge is at
least great as yours.
The format of the typical memorandum is as follows:
TO: John Doe, Manager (Name, Title)
FROM: Sam Public, Engineer (Name, Title)
SUBJECT: Calibration of Water Meter (Descriptive)
Reference: (1) Memo from John Doe dated 7 May 1982
(2) Introduction to Meter Design, A. Jones, Wiley, 1970
Introduction: Briefly explain the significance of what is to be discussed. Tell what was
done and state major conclusions.
Procedure: Describe the nature of the equipment and associated instrumentation. A
sketch is usually helpful and should be considered a requirement. Explain
the test procedures followed and the run sequence and data taken
Calculations: Present equations and procedures used to evaluate the data but there is no
need to derive everything. Show units in your calculations. Mention any
important assumptions. Comment on the expected accuracy of the results.
Results: Discuss the results, which are given in graphical or tabular form later.
Anything that is presented must be mentioned here along with a discussion
of any anomaly r significant features.
Conclusions: Briefly state the significant findings of the study. For example, if your
reason for studying a water meter was to improve test times, you would
conclude that it did or did not actually improve test time.
Illustrative: Present figures, graphs, and tables mentioned in the text. All figures,
graphs, and tables should conform to the instructions given for formal
laboratory reports. The original laboratory data sheet should be included
at the end of this section
Error Analysis
In your report you must point out all possible sources of errors. These errors may take many
forms, such as instrument precision errors, human error, equipment errors, etc. Human errors and
equipment errors are difficult to determine quantitatively, but errors, which may arise from the
precision of measurements, must be identified and accounted for in your report. The following is
an error analysis procedure that can be used to account for these precision errors. This process is
illustrated in more detail in the book (Available at Sterne Library):
Experimental Methods for Engineers, J.P. Holeman, (4
th
, 5
th
, or 6
th
edition), McGraw-Hill Book
company, New York.
Assume that during the experiment, measurements were taken for a number of quantities
designated (x
1
, x
2
, x
3
, , x
n
). It is known that each of these experimental values carries some
experimental error (w
1
, w
2
, w
3
, , w
n
), associated with the precision of the instrumentation,
such that the true value (X) of each measured quantity falls into the range
n n n
w x X
w x X
w x X
t
t
t
M
2 2 2
1 1 1
(1)
Post-experimental calculations require that the experimental results be used to calculate some
desired result or parameter. When this is done, it must be understood that the uncertainty in the
experimental measurements is carried through the post-experiment calculations. This is known
as Propagation of Uncertainty and it must be accounted for in all calculations.
Consider that a parameter is to be calculated based on experimental data such that
R = R(x
1
, x
2
, x
3
, , x
n
) (2)
And the uncertainty of the calculated parameter R based on the uncertainty of the (independent)
experimental measurements needs to be determined. If the uncertainties in all measured
variables are given and represent the true range of error with the same statistical certainty, the
uncertainty in the results is calculated be the relation:
2
1
2 2
2
2
2
1
1 1
1
]
1

,
_

+ +

,
_

,
_

n
n
R
w
x
R
w
x
R
w
x
R
w L (3)
Calculation of equation 3 is best shown through example.
Example 1:
Suppose
3
2 1
x
C
Bx Ax R + + (4)
where (x
1
, x
2
, x
3
) have uncertainties (w
1
, w
2
, w
3
). Then the uncertainty of the parameter R is
given be the equation
2
1
2
3
3
2
2
2
2
1
1 1
1
]
1

,
_

,
_

,
_

w
x
R
w
x
R
w
x
R
w
R
(5)
where
2
3 3
2 1
3 3
3
2 1
2 2
3
2 1
1 1
x
C
x
C
Bx Ax
x x
R
B
x
C
Bx Ax
x x
R
A
x
C
Bx Ax
x x
R

,
_

+ +

,
_

+ +

,
_

+ +

(6)
Such that the uncertainty of the calculated parameter R can be written as
( ) ( )
2
1
2
3 2
3
2
2
2
1
1
1
]
1

,
_

+ + w
x
C
Bw Aw w
R
(7)
This illustrates how each independent variable measurement effects the overall uncertainty of the
calculation. It also indicates how certain measurements can have more or less influence than
others. In this case, if the values of A, B and C are assumed to be of roughly the same
magnitude, then the first two terms represent a constant contribution to the parameter w
R
.
However, the third term is inversely proportional to the measured quantity x
3
, indicating its
contribution can be magnified (x
3
<1) or damped (x
3
>1) depending on the value of x
3
.
A more practical example will also be helpful:
Example 2:
An obstruction-type flow meter is used to measure the mass flow rate of air at low velocities.
The relation describing the flow rate is:
( )
2
1
2 1
1
1
2
1
]
1

p p
RT
p g
CA m
c
& (8)
where the experimentally measured terms are
C the empirical discharge coefficient
A the flow area
p
1
the upstream pressure
p
2
the down stream pressure
T
1
the upstream temperature
The variables R and g
c
are the gas constant for air and the dimensional conversion factor for
gravity, respectively. These are assumed know to a high degree of accuracy.
With the following measurements:
C = 0.92 t 0.005
A = 1.0 in
2
t 0.001 in
2
p
1
= 25 psia t 0.5 psia
p = p1 - p2 = 1.4 psia t 0.005 psia
T
1
= 70
o
F t 2
o
F = 530
o
R t 2
o
R
the uncertainty in the flow rate measurement can be determined assuming the flow rate is a
function of 5 variables:
( )
1 1
, , , , T p p A C f m & (9)
and the uncertainty can be calculated as
2
1
2
1
2 2
1
2 2
1 1
1
1
]
1

,
_

,
_

,
_

,
_

,
_

T p p A c m
w
T
m
w
p
m
w
p
m
w
A
m
w
C
m
w
& & & & &
&
(9)
From the equation for mass flow rate, the derivatives in the above equation are given by:
2
3
1
2
1
1
1
2
1
2
1
1
1
2
1
1
2
1
1 1
2
1
1
1
2
1
1
1
2
5 . 0
2
5 . 0
2
5 . 0
2
2

,
_

,
_

,
_

,
_

,
_

T p
R
p g
CA
T
m
p
RT
p g
CA
p
m
p p
RT
g
CA
p
m
p
RT
p g
C
A
m
p
RT
p g
A
C
m
c
c
c
c
c
&
&
&
&
&
(10)
Taking advantage of the similarity to the original flow rate equation, the calculations of the
uncertainty can be reduced by: 1) substitution of the partial derivatives into the uncertainty
equation (equation 9) and 2) division of the resulting equation by the equation for mass flow rate
(equation 8). The result of this formulation is the percentage uncertainty relation (algebra not
shown, but it is not difficult):
2
1
2
1
2 2
1
2 2
2 2 2
1 1
1
1
]
1

,
_

,
_

,
_

,
_

,
_

T
w
p
w
p
w
A
w
C
w
m
w
T p p
A c m
&
&
(11)
Substitution of the measured values and uncertainties into the above equation results in the
percentage uncertainty for the flow rate calculation
[ ]
% 172 . 1 01172 . 0
10 57 . 3 10 19 . 3 10 0 . 1 10 0 . 1 10 5 . 29
1060
2
8 . 2
005 . 0
50
5 . 0
0 . 1
001 . 0
92 . 0
005 . 0
2 1
6 6 4 6 6
2
1
2 2 2 2 2

+ + + +
1
1
]
1

,
_

,
_

,
_

,
_

,
_

m
w
m
&
&
(12)
Or there is less than a two percent uncertainty in the flow rate calculation. Another point of
interest concerns determining the measurement that contributes the most error. From the above
calculation the largest error is contributed by the measurement in upstream pressure (p
1
). This
error contributes almost all the total uncertainty and should be the most logical place to start in
the improvement of experimental conditions or localization of erroneous results.