home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Hacker Chronicles 2
/
HACKER2.BIN
/
285.WINGREAD.ME
< prev
next >
Wrap
Text File
|
1986-03-21
|
18KB
|
298 lines
Introduction to the WING PROGRAM
For Version 2.05, March, 1986
By Ed Karns for Departure Co.
This is a "free ware" program. See below for details on how to obtain the
latest version of the WING PROGRAM and complete documentation.
THIS IS A SHORT OUTLINE of how to run the WING Program and what may be done
with it to further the evaluation of airframe designs. Although this program
was originally made for use by model sailplane builders the procedures and
results apply to just about any airframe design, powered or unpowered, small
(span down to about 1/3 meter) or large (span up to about ten meters).
Limitations and Liabilities
IN ORDER TO RETAIN the ability to operate on as wide a variety of computers as
possible, certain program abilities had to be sacrificed. The WING PROGRAM does
not deal completely with Canard designs. The Suggested Center of Gravity and
Stability Factor results are NOT accurate for Canard designs. Quite frankly,
we have not been able to find a good algorythm for Canard calculations. We hope
that the users of this program will some day find a good way to obtain these
results and let us know so we can add these features. Note that all other
Canard design results are accurate within the limits of the formulae used by
the program.
With Flying Wing designs the Suggested Vertical Fin and Rudder Area results
will be incorrect. We have been unable to find a workable program for this,
either. However, for virtually any standard airframe with a standard tail
layout located more than about 1/5 span away from and behind the wing's
Geometric Aerodynamic Center Line the results will be satisfactory. Once again,
we would like to see a good program for these calculations.
It would be possible to run a recursive routine to optimize certain results,
were it not for the limitation of Turbo Pascal and Apple Pascal on 8 bit
machines and the memory constraints of most general purpose computers. Some
future version may have these attributes, although improvements to the program
set may come in other areas first.
Some have suggested that it would be nice to have a little graphics to
supplement the results, especially if an optimization routine gets added. We
see nothing wrong with this in a future version. Turbo Pascal has quite good
graphics capabilities and there are easy to use tools for this. It will all
depend on the interest generated from programers and airframe builders. We
would like to give graphics a try in a future version.
THE WING PROGRAM WAS WRITTEN in Turbo Pascal from interpretations of other
programs found in Soar Tech # 1 & 2, published by H. A. Stokely, Virginia
Beach, Virginia and other sources. These other programs originated in such
languages as BASIC, HP-41C language and others. The conversion to Turbo Pascal
achieved a number of advances, the least of which is the ability to run on as
many as 70 % of the world's computers (!) with little or no modification. These
include the IBM PC/XT/AT and all PClones, the Eagle, Kaypro and Osborne
portables and Apple II and Apple III computers that support Apple (UCSD)
Pascal, most Z80 based CP/M machines and many esoteric machines like single and
multi-user 16 bit CompuPros, Molecular multi-user CP/M machines and certain
other mini and mainframes computers. Because we are in the business of dealing
with a variety of machines, versions for all these computers are or will be
available (see below).
An Overview of the WING Program
Acquisition, Protection and Operation
THE WING PROGRAM WAS WRITTEN in the interest of furthering model airframe
design and little, if any, material profit is expected from its distribution.
With this in mind we offer these program files on a "free ware" basis.
IF YOU USE THIS PROGRAM SET AND FIND IT VALUABLE, SEND $ 25.00 TO :
The DEPARTURE COMPANY, 16 JESS AVENUE, PETALUMA, CA. 94952.
In return WE WILL SEND TO YOU the latest version of the WING PROGRAM set and a
reprint of Soar Tech # 2, the fine publication that so much of this program is
based upon. In addition, later versions of this program set will be passed
along, when available, to those who contribute.
WE MUST ASSUME THAT the operator knows how to run his or her computer and he or
she does NOT expect this to be a tutorial in computer operations. We suggest
that you make at least two copy of the files on the distribution disk. The
files included are :
WING.CMD (or WING.COM) ..... The WING PROGRAM, itself.
WING.PAS ............... The main program's Pascal source code.
WINGIT.INC ....... The program's Include File (more source code).
AIRFOIL.DAT .. A file containing data for up to 45 different airfoils.
WINGREAD.ME ............. This information text file.
To run the WING PROGRAM all that is needed are the files WING.CMD (or WING.COM)
and AIRFOIL.DAT.
THE WING PROGRAM IS ARRANGED into a fairly simple, two level, menu controlled
style. The main menu will always reappear after groups of dimensions are
entered or the simulation is run.
* Start the program by typing A>WING <cr>. A full screen menu of choices
appears. The choices include changing units of measure (" u ") as the program
will make all metric / British or British / metric conversions automatically.
* Change the tail fin and rudder dimensions by typing an " f ". You will be
given some questions like "Enter the Fin and Rudder Root Chord <25.00
CentiMeters> >. Go ahead and enter your numbers followed by a carriage return.
Other changes may be made by entering " w " for wing dimension changes and " s
" for horizontal stabilizer dimension changes. NOTE that entering letters in
place of numbers will cause the program to bomb.
* Enter important stuff like tail fin height and chords, etc. If the numbers
given are acceptable, simply enter a carriage return <cr> and the dimensions
will remain the same. The program comes equipped with a set of default
dimensions and weights for an F3B, high performance, model sailplane with an
Eppler 387 airfoil. If none of the dimensions are changed, these default values
will be used in the simulation. The program has a reset choice on the main
menu. Using this option (" r ") returns all the values to their original, F3B
settings.
* After the questions have been answered, enter " d " to display calculated
information like suggested wing dihedral angle and center of gravity locations,
etc. Then enter " t " to display important tail fin, rudder and stabilizer
information like tail area, stabilizer area, etc. Note that the " d " and " t "
entries return primarily fixed dimension information like wing and tail areas,
geometric aerodynamic center line locations and suggested changes to improve
airframe aerodynamic stability. The " x " is used to run the flight simulation.
* If you are satisfied with your entries, then pick an airfoil with " a ".
Notice that only the first dozen airfoil choices have foil names. These are
supplied with the program and are part of the file AIRFOIL.DAT. The other
"Undefined" airfoils are for you to modify with the Editor section of the
program. Pick one of these airfoils by entering its appropriate number. A
carriage return only automatically chooses the Eppler 387 airfoil, # 1.
* Now run the flight simulation with the " x " choice. What you get is a
complex set of columns of numbers. These represent Horizontal and Vertical
Velocities and corresponding Lift over Drag Ratios in relation to Coefficients
of lift. The displayed results for each set of calculation are the Reynolds
Number (top center), Air Speed (the two numbers at top right), Total
Coefficient of Drag (bottom center), Percentage of the Wing's contribution to
the Total Coefficient of Drag (bottom right) and the information in the middle.
This middle information is a record of Horizontal Velocity (VX), Sinking or
Vertical Velocity (VY), Lift/Drag Ratio (L/D) in relation to each Coefficient
of Lift (CL) from 0.00 to 1.24 in increments of 0.02. To arrive at these
numbers, the central processor inside your computer must perform the following
calculations :
(* Note : Lift_Coef is a given. Sqr means Square. Sqrt means Square Root. *)
(* Aspect_Ratio, Atmos_Density, Wing_Area, etc. are recalculated each time *)
(* a dimension or weight is changed in another part of the program. *)
Induced_Drag_Coef := (( 1 + ( 0.03 * Aspect_Ratio )) *
( Sqr ( Lift_Coef ))) / ( Pi * Aspect_Ratio );
Parasite_Drag_Coef := 0.005 + ( 0.004 * ( Sqr ( Lift_Coef )));
Total_Drag_Coef :=
Profile_Drag_Coef + Induced_Drag_Coef
+ Parasite_Drag_Coef + Tail_Drag_Coef;
Air_Force_Coef := Sqrt ( Sqr ( Lift_Coef ) + Sqr (Total_Drag_Coef ));
Speed := Sqrt (( ( Weight + Ballast_Weight ) * 2 ) /
( Atmos_Density * Wing_Area * Air_Force_Coef ));
Vert_Speed := ( Speed * Total_Drag_Coef ) / Air_Force_Coef;
Horiz_Speed := ( Speed * Lift_Coef ) / Air_Force_Coef;
Glide_Ratio := Horiz_Speed / Vert_Speed;
Reynolds := FRE * Mean_Chord * Speed ; (* FRE := 68459.6 *)
Tail_Drag_Coef := (( 3.75615 / Sqrt ( Reynolds )) / Speed ) + 0.00850;
(* Above from Soar Tech # 1 article and TI 59 program by Martin Simons *)
(* These last two formulae are used on the reiteration of calculations *)
* In addition, six comparisons are made to determine dynamic changes from
previous calculations so that the program can announce Stall Conditions and
show dynamic changes in Sinking Velocity, Horizontal velocity, Lift/Drag and
Drag Coefficient Percentage. All these calculations are made each time a new
group of figures are displayed !! That's more than 1000 calculations for each
full screen simulation run !! Our machine does this in less than ten seconds
(with no stall conditions). Note that there are changes in highlighting of the
displayed result record. These highlighting changes represent changes in
increasing or decreasing figures (reversals of changes !) :
** VX, VY and L/D changes from dark to light or light to dark means VX,
VY or L/D changes from increasing values to decreasing values or
decreasing to increasing values. Example :
L/D
< 16.66 > light, increasing
< 16.68 > light, increasing
> 16.67 < dark, decreasing
> 16.65 < dark, decreasing
< 16.69 > light increasing ...
** CL changes from dark to light or light to dark mean changes from
increasing to decreasing values of the Wing's Coefficient of Drag
Percentage of Total Drag Coefficient ! Example (as above).
* Stall Conditions may appear toward the end of a simulation run. These
represent the point at which valid airfoil data comes to an end. In real wind
tunnel tests the last result points on the graphs are the last valid result
points. Any simulation results beyond the last of these points will be in
error. The simulation reports this with the stall condition sign. Note that if
there is a dramatic difference between two L/D data points and CL is greater
than 0.4 then the stall condition sign will appear. This represents a genuine
stall condition ! In order to have an airframe simulation that performs with
your particular airfoil, you will end up :
Using the Airfoil Editor
THERE ARE THREE IMPORTANT things to remember when using the Airfoil Editor.
First, DO NOT EDIT the files on your master disk. Run the Airfoil Editor on
copies of the program files only ! The second thing is Always Enter a number at
the beginning (where CL = 0.00) of each page of Coefficient of Lift verses
Coefficient of Drag Data (hereafter called CL/CD Data). This second rule may be
ignored if you do not mind dealing with incorrect results during the first part
of the simulation run or in the special case where there are only three pages
of data and the fourth page has to be "zeroed out". The third rule is, whenever
possible, enter Real Wind Tunnel results rather than theoretical projections.
Even though the third rule is more of a suggestion than a rule, the truth is
that calculated airfoil data is almost always dramatically less accurate than
Real Wind Tunnel data. When theoretical or calculated data is as little as 3 %
in error from real data, the simulation run may produce errors in excess of 20
% !! As in most other things in life, there's no substitute for the real thing.
A sample entry into the Airfoil Editor might go something like this. Since the
Editor is part of the main program some niceties were sacrificed in the
interest of memory space. It has been suggested that the Editor might be better
separated from the main program so that the main program could perform some
optimization routines and we may very well do this in a later version. For now
the Airfoil Editor seems to work OK as is. One of the less convenient things
about the Editor is that you can't backup (return to a previous entry). This
"feature" will become apparent as you use the Editor. For now enter " e " for
the Editor and you will be given the list of 45 airfoil choices. Choose one of
the airfoils marked "Undefined". We will change it to a useful set of data. In
reality all of the airfoils except the Eppler 387 may be modified and we would
encourage you to do so if you find better wind tunnel data or wish to use the
space for your own special foils. If you make a mistake (nothing is perfect)
and mess up the data file, you can always copy the data file, AIRFOIL.DAT back
onto your work disk. If you do "trash" your master disk just send $10.00 with
the old disk (or any working blank disk) to : Departure Company, 16 Jess Ave.
Petaluma, Ca 94952.
You will be asked to enter the airfoil name. Enter "SPECIAL FOIL #2" or what
ever. A " c " after the name represents that the foil information is from
calculations, not from real wind tunnel data. After the carriage return <cr>,
you can enter the Reynolds Numbers for each of four pages of data, one Reynolds
Number for each page. If you have only three pages of data to enter (as in this
case) then you should enter a very high Reynolds Number for the fourth page. In
this case we will use 10000000 (that's ten million) for the fourth Reynolds
Number. The WING PROGRAM has a method that will make its own tables of data
based on interpolations of the data that you enter. So when you make the data
page for Reynolds Number = 10000000 the program will effectively ignore the
data on this page as few model airframes will fly above R # = 3 or 4 million.
The calculations for a Reynolds Number above 10 milloin will be ineffective and
ignored by the simulation.
A fresh page of CL/CD data points will be presented. !! IMPORTANT !! You should
enter a number for the first CD entry (see above three rules). If you enter any
number equal to or greater than one (1) at any CL/CD data point, the page of
CL/CD data will be accepted and the next page will be presented. This is handy
if you wise to change only a few points of data on one or two of the pages.
Enter only those points that you know are valid. The program fills in the
points in between mathematically. The program effectively "draws" a straight
line of points between each valid entered point. If you have valid wind tunnel
data be sure to enter only those points that actually represent valid wind
tunnel results. It is best if you not try to "smooth out" any data curves by
entering "estimated" points. Remember, "garbage in = garbage out". The program
will usually do a much better job of interpolation than you can by eye from a
graph. Those interested in program construction should examine the source code
(WING.PAS) for a good idea of how this works.
The fourth and last page of data is a special case. If you have valid CL/CD
data for this page, fine, use it. If not, the idea is to make the last page
ineffective to the calculations and rely on the program's ability to
interpolate CL/CD data points from the other pages. This is done by effectively
"zeroing out" this last page. To do this you should have used a very high
Reynolds Number, a Number that your airframe will never "fly at". If your plane
weight is reasonable and your airfoil data is close to "real" then a Reynolds
Number above 10000000 (ten million) or more will be out of reach for your model
airframe. To further "zero out" the last page you should enter a very low
number for the very last CL/CD data point, say 0.000001. All other entries for
this page, INCLUDING THE FIRST CL?CD DATA POINT (CL = 0.00), should be left at
0.0. Note that any point may be "zeroed out" by entering 0 <cr> at that point.
The Editor automatically saves the new foil name, Reynolds Numbers and data
points after the fourth page is done. This is to prevent program operation
problems like open files, loose ends and lost end of file marks that can
"glitch" a file. If any data points are in error, simply re-edit the foil.
THE WING PROGRAM REPRESENTS an experiment for us. We have hopes of producing
improved versions of this program set as well as other complete programs of
this type. We may not bother if responses to our "free ware" offer are not
forth coming.
We are very interested in your comments. Please address comments to :
Ed Karns
c/o Departure Company
16 Jess Avenue
Petaluma, California 94952