home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Mac Mania 5
/
MacMania 5.toast
/
/
Tools&Utilities
/
Plotfoil 3.2
/
naca-1.0
/
ReadMe
< prev
next >
Wrap
Text File
|
1995-09-13
|
6KB
|
187 lines
NACA -- a program and library to generate NACA Airfoil Ordinates
By: S.E. Norris
norris@cfd.mech.unsw.edu.au
Introduction:
In the 30's and 40's the NACA undertook a systematic investigation into
aerofoil design (or rather airfoil, since they were American). The airfoil
sections studied are widely used since a large body of experimental data
for the sections has been published. Now, barely a week goes by without
someone asking for the section coordinates in sci.aeronautics. Hopefully
these people might get some help from this code.
Naca is a utility to generate NACA section ordinates. Most of the foils
in Abbott and von Doenhoff[1] can be generated, with the exception of the
7-series foils. Thus it will produce 4 and 5 digit foils (original and
modified), the 1-series sections, and the 6-series sections (with uniform
and partial loading, and with the original and A series profiles).
The code to generate the airfoil section is included as a library. Thus
it can be linked to your code, with only one call being neede to generate
any section.
You should be able to find the code at
ftp://cfd.mech.unsw.edu.au/pub/src/naca-1.0.shar.Z
The Naca Utility:
The naca utility is run with the command line arguments
naca NACA_Number N
where N is the number of points defined around the foil. Actually N+1
points are spat out of the program since the trailing edge point is
dumped out twice. Thus the command
naca 2412-64 100
would write 101 points from a modified 2412 section to stdout.
In general the foil numbers are identical with those found in
Abbot and von Doenhoff. Examples are
0012 Symmetric 4 digit foil, 12% thickness
23010 Cambered 5 digit foil, 12% thickness
2402-64 Modified cambered 4 digit foil.
23011-32 Modified 5 digit foil.
16-012 Symmetric 1-series foil.
The naming convention for the 6-series foils is rather more complicated.
The first two numbers are followed by a subscript giving the width of
the drag bucket in terms of lift coefficent in tenths. This information
is not needed to define a section and is thus ignored. Thus a
64 -018
3
is generated by the command
naca 64-018 100
A later series of thickness distributions has a 'A' instead of the dash,
and can be generated too
naca 64A018 100
If a foil thickness profile is generated by linearly increasing/decreasing
one of the originally derived thickness distributions, then the thickness
of the original section is given in the original NACA number system by a
bracketed subscript (with the width of the drag bucket leading the subscript).
Thus a 17% thick section can be generated from a published 18% thickness
distribution
64 -217
(318)
These sections can be generated by naca; however the drag bucket number is
dropped when feeding in the number (and remember to put quotes around the
number when using brackets on the command line)
naca "64(18)-217" 100
Since there is no 17% thickness section, if the (18) is dropped from the
number the naca program will interpolate from the nearest section, and
will remind you of this interpolation in the output. Thus the first line
of the output from the
naca 64-217 100
command is
# NACA 64(18)-217
telling you that the code had to interpolate from an 18% section.
Finally, the mean line can be modified from the original serction by use
of the trailing a notation:
naca 64-217a=0.5
will generate a NACA 64(18)-217 a=0.5 section.
The points are distributed acording to a cosine rule (see[2] p132). This
puts more points at the ends of the foil, thinning out the distribution
in the foil's midsection.
The NACA library:
The NACA foils can be generated via a subroutine call. The name of the
foil is passed in as a string, and the section data is returned. Look in
the naca.f file for an example of the call.
Yuck, what horrible code!
I wrote the library back in 1992/3 when I was still learning the ropes at
the Uni of Auckland. The code is pretty horrid, yes, but it still works,
so don't fix it. It orginal hung off the bottom of the GUI to my 2D aero
program, so it is reasonably robust.
Porting:
I have tested the code on IBM RS6000's, SGI Indegos, Indego2's, and a
personal Iris, a Cray YMP-EL, the frontend of a CM5, a Sun, some
HP 735's, DEC AXP's running OSF, and me braindead Linux box. Hopefully
after that the code is reasonably portable. The only system calls are
in naca.f: these are either getting command line args (iargc() and getarg()),
or the exit() call (prevents many compilers from throwing an ugly STOP at
you). Apart from that it is pretty much plain DoD Fortran.
Instalation:
Read the Installation document.
To Do:
A man page should be made.
Other distributions of points apart from a cosine distribution should be
available.
The NACA 1- and 6-series profile data should be hard coded into the program,
rather than loaded from file.
Bugs/Problems:
The astute might notice that the NACA 64A010 foil isn't as written in
Abbott and von Doenhoff -- the 7.5% station is 2.805 rather than the
published 2.905. This was done to ensure fairness in the foil -- I
assume that there is a misprint in the book at this point.
The U77 libraries on the HP's that I have access to are miscompiled it
seems. Whilst most functions in the library have a trailing underscore,
the exit() function doesn't, and so can't be linked with. Rather than
doing a whole lot of buggering around adding underscores to the calls to
the other functions in the library, I have included a dummy exit() routine
with the code. This is also used in the Linux port, since the f2c libraries
don't seem to have this routine either.
References:
[1] Abbott, I. H. and von Doenhoff, A. E., "Theory of Wing Sections,
Including a Summary of Airfoil Data", McGraw-Hill, 1949.
Reprinted by Dover.
[2] Moran, J., "An Introduction to Theoretical and Computational
Aerodynamics",Wiley, 1984.
----------------------------------------------------------------------------
Stuart Norris |
Email: norris@cfd.mech.unsw.edu.au | School of Mechanical Engineering
Work: Tel: (+61 2) 385 5738 | University of New South Wales
Fax: (+61 2) 663 1222 | Sydney 2052
Home: Tel: (+61 2) 313 0662 | Australia
----------------------------------------------------------------------------