Manual for Technical Report Writing

Department of Aerospace Engineering & Engineering Mechanics

Technical Report Writing
In addition to performing the engineering activities, an engineer has to
prepare operating and assembly manuals, requests for proposals (RFP),
responses to proposals, research and development reports, and articles for
publications. Writing for these specific purposes is called technical riting.
!echnical riting is totally different from creative riting. !he difference
beteen the to styles of riting is similar to the difference beteen a te"t
on Physics and a novel. # technical report or a paper generally consists of
reporting and$or interpretation of data$physical phenomena. !he report
should be ritten in a purely ob%ective and impersonal manner and reflect
the riter&s effort in addressing a technical problem. !he results are to be
%udged on the basis of engineering e"perience and common sense. 'ost
technical reports are ritten in third person, past tense. (enerous use of the
technical terminology is employed. For the description of the laboratory
equipment, scientific las and the presentation of results, present tense
should be employed. !he use of personal pronouns should be avoided. #
technical report generally contains a clear outline of the problem,
mathematical equations, draings, charts, tables and graphs, etc. For
completeness purposes, they alays require e"planations at appropriate
locations ithin the te"t of the report. !echnical reports must be accurate
and to the point. )uperfluous and unsupported assertions should never be
made. *onclusions and recommendations must be based on the results
presented in the report. It must be technically correct and ithout any
ambiguities. In this respect technical report riting is hard or+.
# report or a technical paper usually consists of the folloing parts.
,. !itle Page
-. .omenclature
/. #bstract (or )ummary)
0. Introduction (or 1ac+ground)
2. !echnical #pproach (or 'ethodology)
3. 4escription of 5"perimental )etup and !ools 6sed
7. Results
8. *onclusions
Explanation
1. Title Page:
!he title page should contain the title of the pro%ect, the
authorship and their affiliation and date of submission.
2. omenclature:
!his part of the report is optional. 9ere all the symbols used
should be defined. #lternately, the symbols can be defined in the te"t here
they appear for the first time. For big and comple" reports (li+e a big
proposal or final report on a contract), it is suggested that a list of the
symbols ith appropriate definitions should be provided up front. For
reports ith limited ob%ectives (e.g. research papers or class reports), it is
recommended that the symbols should be defined ithin the te"t after they
appear for the first time.
!. A"stract:
It is a brief summary of the essential content of the technical
report. Its length should be about -:: ords or less. !he abstract usually
contains a brief statement of the ob%ectives, the methods used to fulfill those
ob%ectives, most significant results, and conclusion. .o graph, s+etch, table,
or equations should be used or referred to in the abstract. It is recommended
that the abstract be ritten after the report has been finished. It is imperative
that a great deal of attention is paid to the riting of the abstract. # ma%ority
of the readership (li+e upper administration) of the report ill confine their
focus to this part of the report. ;nly those dealing directly ith the sub%ect
matter of the report ill loo+ into the details.
#. $ntro%uction:
!he purpose of the introduction is to provide the appropriate
bac+ground on the sub%ect matter. !he introduction should describe the
ob%ectives of the or+ and the method and scope of the investigation, and
the reasons for carrying out the or+. #ppropriate references to any earlier
or+ must be ac+noledged. 6sually, figures, s+etches, or tables are not
presented in the introduction. 6nli+e the abstract, there is no upper limit on
the length of the introduction< hoever, an effort to +eep it concise is alays
recommended.
&. Metho%olog' or Technical Approach:
!his part of the report should identify the theoretical principles
and equations used in accomplishing the or+. !he equations should be
numbered and follo the order of the principles they represent. #ny
supporting s+etch of diagram related to the problem should be provided.
4efinitely, these are not the diagrams representing the results that you have
obtained. !hese diagrams must be labeled and should appear after a
reference has been made in the te"t.
# discussion of the sources and the e"tent of e"perimental
and$or computational uncertainties should be included. ;ne must be
cogni=ant of the fact that the precision of the measuring device is compatible
ith the ob%ectives of the investigation. )imilarly, the computational
technique that is being used also is stable and does not pollute the results.
For e"ample, a scale usually employed in grocery stores cannot accurately
measure a microgram of an e"pensive chemical. # careless inversion of an
ill>conditioned matri" can affect the final outcome of a computational
e"periment.
(. Description of the Experimental )etup or Tools *se%:
# s+etch of the setup ith appropriate identification of the
components should be included. #ll the measuring devices and the samples
tested should be briefly described. In case it is purely a computational
effort, a description of the numerical methodology or the use of speciali=ed
softare and the language in hich the computer code is ritten should be
provided.
+. Results an% Discussion:
!his is the most important part of the report, here the findings
have to be reported. 'ost of the engineering results are presented in the
form of tables, charts, and graphs. 5ach of these entities should be
numbered and provided ith a one>line caption. !hese entities should then
be fully described in the te"t as to hat each table and chart stands for and
hy it is important that it has been included in the report. #ppropriate
discussion of the results for trends, self>consistency, accuracy, and meeting
the original ob%ectives must be provided. It is the duty of the riter to sort
out the results in a logical and systematic fashion and convince the reader
about their validity. ?uantitative discussion of any errors from the e"pected
results should be clearly pointed out. #void any discussion of irrelevant and
insignificant aspect of the results. (.ote@ proposals do not have results
sections.)
,. -onclusions:
!he most significant conclusions reached in the results and
discussion section should be summari=ed herein. !hese should be listed in
the order of their significance and preferably in an itemi=ed form.
References to figures or tables should not be included, and no ne
information should be introduced in this section. 9oever, suggestion for
further investigations can be made.
)amples of Technical Reports
!hree samples are given here< the first is suitable for engineering students
riting a homeor+ report, the second is a technical paper, and the third is a
technical presentation.
, A to be sent (homeor+)
- A !echnical Paper@ #nalysis of Flo in *onverging>4iverging .o==le,
6niversity of *incinnati, .ovember -::-
a. omenclature
A(x) area variation
f function
k
inlet
constant
k
exit
constant
M Mach number
ratio of specific heats of the gas flowing through
the nozzle
x distance along the nozzle
tol tolerance for numerical solutions
Subscripts:
inlet refers to the nozzle inlet
exit refers to the nozzle exit
Superscripts:
* throat area
b. ;b%ective
!he main ob%ective of this pro%ect is to solve for the 'ach numbers
(subsonic and supersonic) corresponding to each value of " in a
converging>diverging no==le and to plot 'ach number vs. ".
!he area #(") variation for the no==le ith respect to the length "
is given by the folloing e"pressions@

A(x) = 1+(
+
inlet
A,)(,>"$2)
-

0 x 5

A(x) = 1+(
+
e"it
A,)

(x 5)
2
xexit 5

5 x 10
(,)
here +
inlet
B +
e"it
B /, and "
e"it
B ,:.
!he 'ach number distribution of gas in the no==le is governed by
the folloing equation@

f (M,A) =
1
M
2
2
+1
1+
1
2
M
2

+1
1

A
A
*






2
= 0 (-)
here #
C
B !hroat #rea B ,.
c. 'ethodology
!he relationship beteen area and 'ach number comes from gas
dynamics and is given as@

f (M,A) =
1
M
2
2
+1
1+
1
2
M
2

+1
1

A
A
*






2
= 0 (-)
here A is the area at any section of the no==le. #
C
is the throat
(minimum area B ,) area and is the ratio of specific heats of the
gas floing through the no==le. For air, B ,.0, and for realistic
situations arising in propulsion applications, can be closer to
unity.
!he governing flo equation cannot be solved analytically, so in
order to analy=e the given problem it is solved numerically using
.eton&s method. We initiate one loop on ", hich increases "
from : to ,:. It can be observed that for a particular value of " e
can obtain a corresponding value of A from equation (,) that leads
to one equation for '. !he problem is reduced to solving this
equation using .eton&s method. .eton&s method requires an
initial guess for '. For a guess that is closer to the solution,
.eton&s method ill converge quadratically. For a poor or bad
guess, the .eton&s method can diverge. !he iterative solution
process that updates the solution is given as follos@

Mk + 1 = Mk
f (Mk,A)
(
f
M
)M = Mk
+ B :, ,, -D, here

f
M





=
2
M
2
+1
1+
1
2
M
2

2
1

2
M
3






2
+1
1+
1
2
M
2

+1
1
#s it is +non that at any station " there is a possibility of to
solutions, supersonic and subsonic, the choice of the initial guess
further becomes critical.
!he stopping criteria for iterations is '
+E,
>'
+

tol, here
+ is an iteration inde" and value of tol (tolerance) is decided by the
user and represents the level of accuracy of the final solution.
;nce the convergence is attained, the final 'ach numbers and the
corresponding values of all " are stored in an array and plotted.
d. Results
# computer code for the problem has been ritten in 'atFab and
is attached as #ppendi" ,. !he results of this computation are
shon in Figures , and -. !he variation of area of the no==le
versus " is displayed in Figure ,. *learly the throat area of the
no==le is one and is located at " B :. !he 'ach number
distribution along the no==le is shon in Figure -.
It may be noticed that in the converging portion of the no==le, a
subsonic 'ach number at the inlet increases to , at the no==le
throat, hereas a supersonic 'ach number at the inlet decreases to
, at the throat. )ince the deceleration of supersonic flos ta+es
place across a normal shoc+ ave, this branch of the solution
though computed is non>physical. 9oever, in the diverging
section of the no==le both subsonic and supersonic solutions are
possible. !he reali=ation of either one is determined by other
conditions. !he convergence of the .eton&s method requires that
appropriate initial guess is used to initiate the iterative process. It
as also observed that different initial guesses are required to
ma+e the subsonic solution converge in the converging (for "<2)
and diverging (for ">2) of the no==le. For "<2 the solution
converged if e started ith an initial guess of ' B :., hile for
">2 the subsonic solution converged if e started ith ' B :.:2.
.o such thing as observed hile running the code for supersonic
case, the supersonic solution converged if initial guess given is
greater than /.
e. *onclusions
'ach number distribution in a converging>diverging no==le has
been calculated using .eton&s method for subsonic and
supersonic branches. .eton&s method converges quite rapidly
and the convergence depends upon the initial guess.
Appen%ix 1
'atFab *omputer *ode (G solution of flo in a converging>diverging
no==le using .eton&s method)
function [M,y]=Newton_3(fun,fun_pr,M1,tol,max)
% Find zero near x1 using Newtons method
% Input:
% fun string containing name of function
% fun_pr name of derivative of function
% fun_2pr name of 2nd derivative of function
% x1 starting estimate
% tol allowable tolerance in computed zero
% max maximum number of iterations
% Output:
% x vector of approximations to zero
% y vector of function values, fun(x)
newton_1_data
counter = 0;
for x = 0:.1:5;
counter = counter + 1;
i = 0;
iter = 0;
M(1) = M1;
y(1) = feval(fun,M(1),x);
y_pr(1) = feval(fun_pr,M(1));
for i = 2:max
M(i) = M(i-1) y(i-1)/y_pr(i-1);
y(i) = feval(fun,M(i),x);
y_pr(i) = feval(fun_pr,M(i));
if abs(M(i) M(i-1)) < tol
%disp(Newton method has
converged); break;
end
iter = 1;
end
if (iter >= max)
disp(zero not found to desired
tolerance);
end
n = length(M);
k = 1:1:n;
out = [k M y];
Mach_num(counter) = M(n);
end
out = [Mach_num};
x = [0:.1:5];
plot(x,Mach_num);
disp(out)
hold on;
function y = my_func(M,x)
ki = 3;
y = (M.^-2)*((5/6) +(1/6)*M.^2).^6 -(1 + (ki
-1)*(1-x/5)).^2).^2;
function y = my_func_pr(M)
y = (-2/ (M^3)) * (5/6 + (M^2)/6)^6 + (6/(M^2))
* (M/3) * (5/6 + (M^2)/6)^5;
%y = (2.4*M.^-l)*((5/6)+(1/6)*M.^2).^5 -(2*M.^-
3)*((5/6)+(1/5)*M.^2).^6;
/ A !echnical Presentation
Poer Point presentation@
,. !itle page ith the authors& names and affiliation
-. ;utline of the presentation
Problem statement and bac+ground
Formulation of the problem
'ethod of solution
Results and discussions
*onclusions
#ll slides should be brief and simple.