FDE's flap lift coefficient is very important
because, for the first time, it lets people set the stall speed
with flaps correctly. I changed the stall speed with flaps of
one of my aircraft from 75 kts to 70 kts and it tested correctly
with the stall horn sounding at 71 kts. To set the flap lift
coefficient to achieve the proper stall speed, use the following
CLF is flap lift coefficient
W is gross weight (lbs)
S is wing area (sq feet)
VSC is stall speed clear (kts)
VSF is stall speed with flaps (kts)
CLF = 294.46 * (W/S)*(VSF^-2 - VSC ^-2) (using
BASIC notation for reciprocal of square of velocities)
Thanks to Tom Goodrick at firstname.lastname@example.org
for this formula.
Tom also commented about another variable
which is missing from the initial release of FDE. This variable
affects the rate of fuel burn. For piston engines, this value
is usually 138. He suggests that for most aircraft, it be adjusted
downward to the range of 124 to 130, although he has a T28 aircraft
which is set to 90.
NOTE: Download the entire package again to
use the most recent changes to both FDE and the control file.
When you run FDE, the new variable appears immediately after
"Maximum Power (horsepower - each)." (For those interested,
this variable in in Section 500, Offset 32).
Thanks again to Tom. If you're interested,
Tom's credentials are impressive. He's an aeronautical engineer
and a pilot for more than a "few" years.
This updated control file is now incorporated
if you download FDE.
I was glad to see that FDE has a few things
like spoiler lift and spoiler pitch coefficients and flap lift
coefficient. I just used it to correct the spoiler pitch on the
original Lear 45. They had it at -.1 but it should be +.1 giving
a nose-down pitch with spoiler deployment.
Thanks a third time to Tom Goodrick.
I have found that setting the EGT value to
.5 (FS default is 1.0) provides a more accurate reading when
adjusting the mixture control in piston engine aircraft.
Thanks to Bob Langendorfer for this report.
I recently discovered what an "Unknown"
variable does: 329-08 is the autothrottle variable. 0 is disabled;
1 is enabled. I checked this with two aircraft.
This information has been added to the control
file and is now updated in the download.
Thank to Dale Johnson for this report.
FLIGHT SIM AIRCRAFT
Several users have been testing FDE with CFS
aircraft. Among them is Mike Bonomo and his "Old Geezer"
squadron. He reports that he's now able to accurately model many
of their war planes' flight characteristics.
If you have any questions regarding their
activites please check out the Old Geezer site at http://www.flightsimmers.net/oldgeezers/index.html.
WING MEAN CHORD
You can determine the main wing mean chord
value as follows:
MC is mean chord value
WA is main wing area (sq. feet)
WS is main wing span (inches)
MC = WA * 144 / WS
These values are found in Section 1204
Thanks to Rabbijah Guder email@example.com
for this formula.
You can determine the fuselage angle as follows:
FA is fuselage angle (degrees)
NGPV is Nose/Tail Gear Position Above+/Below- of C of G (inches)
MGPV is Main Gear Position Above+/Below- C of G (inches)
NGPL is Nose/Tail Gear Position Fore+/Aft- of C of G (inches)
MGPL Main Gear Position Fore+/Aft- C of G (inches)
FA = - arctan( (NGPV - MGPV) / (NGPL - MGPL)
) NOTE: Minus sign
The value of FA is found in Section 301 and
the other values are found in Section 1004.
Thanks to Rabbijah Guder again for this information.
OF GRAVITY ABOVE GROUND
Another discovery by Rabbijah Guder is a forumla
to determine the COG above Ground (Section 301-6):
C of G Above Ground = -(LBG * TAN(-FA) + HBG)
- HBG is Main Gear Position Above+/Below- C
of G (Section 1004-12) unless it's a taildragger in which case
HBG is Nose/Tail Gear Position Above+/Below C of G (Section 1004-24)
- LBG is Main Gear Position Fore+/Aft- C OF
G (Section 1004-16) unless it's a taildragger in which case LBG
is Nose/Tail Gear Position Fore+/Aft- C of G (Section 1004-28)
- FA is fuselage angle (Section 301-10)
AILERON, RUDDER FINE CONTROL
Rabbijah Guder believes that the tables in
Sections 341, 342 and 343 are used to fine adjust the elevator,
aileron and rudder controls. Here's his description:
For CFS, each table contains
7 pairs of values which describe a non-linear function. The
first value represents how much the control was moved by the
pilot with the joystick ranging from 0 (neutral) to 1 (maximum).
The second value is a factor that is multiplied
by the respective control factor e.g. Elevator Control Factor
For this example, assume that the Elevator
Control Factor has a value of -1000.
Moving the joystick through 50% (.500000)
of its range, you'd expect the effective force throuh the elevator
to be .500000 x -1000 = -500
If the table has a pair -.500000, 0.875 then
turning the joystick to 50% (.500000) yields a different value
this being 0.875 x .500000 x -1000 = -437.5
What can this be used for? Looking at the
stock P-51D in CFS, you can see that the elevator is more effective
if turned left, therefore left turns are faster. Furthermore,
this can be used to adjust the sensitivity of joystick movement
shortly before stall.
Section 342 is used similarly for the aileron
and Section 343 for the rudder.
Rabbijan Guder believes that 3 variables affect
the "stiffness" of the elevator, aileron and rudder
(320-00, 320-08 and 320-16)
He says, "They don't seem to effect the
flight dynamics, but they completely change the feeling (how
it handles) of the flight expereince."
Setting these to a high value makes the control
behave very "hard" - needing more force. Additionally
the aircraft tends to make small oscillations near its top speed
in the pitch direction. This is the reason I name them stiffness
and not "laziness".
What can this be used for? For business or
civil jets I suggest setting these values low since all of them
use servo controls (autopilot). For WWII planes I suggest settin
them high since they used cables to connect the controls to the
flight stick. This brings more realisim.
Mosselman has pointed out that using
CFS, the B17g has an aircraft type of 2 (Section 105). Previously
we called this field "Helicoptor?" and believed that
it was a boolean type. The identified types are: 0=normal aircraft,
1=helicoptor, 2=CFS AI aircraft.
Additionally, he tells us that the CFS V1
aircraft has an engine type of 4 (Section 310) for "rocket".
The identified types are: 0=prop, 1=jet, 2=glider, 3=helicoptor.
He's also been experimenting with Section
300 which he says are mentioned in the FS98 S.D.K. - control
Jan corrected us with Section 1004 landing
gear types - 0-fixed, 1=retractable, 2=skid, floats, 4=skis also
gleaned from the S.D.K.
ENGINE - MANY NEW VARIABLES
Thanks to Guntram
Strasser of Technic Direct. Gunti has provided us a very
complete mapping of the variables in Section 500 for aircraft
that have reciprocating engines. If the date on your control
file (fdectrl.txt) is 03/18/99 or later, the control file includes
these new variables.
Thanks to Samuel
Schmid for various additions to our collection of variables.
He's also challenging others to experiment with Sections 1400-1404
which are helicoptor related variables.
Mike Eustace, has asked us to add 'Unknowns'
to the control file so that designers can experiment. Warning:
we have had no experience in identifying or modifying these variables.
There's a good likihood that some fo these experimental control
file definitions are inaccurate.
If you download FDE.ZIP (after of 3/13/99),
the control file includes several verified flight variables.
These include the yaw and roll centers and vertical and lateral
inertias. Thanks to Rabbijah Guder who identifed these critical
Thanks to Rabbi and Dietger of Germany who
provided the description of Section 505 variables which are specifically
for CFS reciprocating engines.
FLAP CYCLE TIME
Rabbi also "discovered" the flap
cycle time in Section 1101, Offset 12.
STABILITY FACTOR #2 - SMOOTHS DIVERGENCE
I discovered the cause of aircraft that exhibit
divergent oscillation when flown using the autopilot. The Beech
Duke and the Aerostar 600A are two examples. I removed the Aerostar
(an original design) from my web site because of the divergence.
After two days of testing I found the solution. For planes of
this type and size (medium twins), the "Pitch Stability
Factor #2" must have a value of about -40000 which is what
the Piper Navajo had and it worked OK. The Duke had a value of
-250000 and had a very strong divergence. It would go out of
control in about a minute. I gave it -40000 and it smoothed right
out. The Aerostar had a value of -70000 and a mild divergence.
It would go out of control only after several minutes. The value
-40000 straightened it right out. A common feature of these planes
is that they are very sleek with high power. Their cruise speed
is at least 2.3 times their clean stall speed. It would sure
be interesting to know how this parameter is calculated for the
This note from Tom Goodrick.
OF INTERTIA - A PRIMER
Probably the most esoteric aspect of setting
up an aircraft model for FS98 is the estimation of Moments of
Inertia or MOI's. You need to enter an MOI for roll, pitch and
yaw referenced to the three standard aircraft axes. In theory,
these are calculated by summing the product of the mass of each
component with the square of its position radius about the axis.
This is a very daunting task. Yet the MOI's play a very important
part in the flight characteristics of the aircraft. The roll
MOI determines how much the aircraft resists the roll control
input, for example. Also, the relative values for the different
pairs of axes determines the coupling between axes. A simple
roll input will typically produce secondary yaw and pitch rotations
as well as a roll rotation. The nature of this effect - indeed
the direction of rotations - is determined by the difference
between the pairs of MOI's: (Ix-Iy), (Iz-Ix), (Iy-Iz). Hence,
to model a plane's behavior well, it is important to nail down
these MOI's with reasonable accuracy.
Dr Jan Roskam, has published an 8-volume set
of texts on aircraft design. He has taught courses in aircraft
design for some time at the University of Kansas. He came up
with a method for estimating MOI's that uses only basic aircraft
weight and dimensions combined with a set of coefficients of
radii of gyration which Dr Roskam has compiled for several types
of aircraft. We simply determine which type suits the design
we are concerned with, take the coefficients for each axis and
the approriate dimension and calculate MOI's. The formulae are:
I first started using this with Flight Shop
for FS5.1 because I noticed that the estimation method used within
Flight Shop was way off making many aircraft behave strangely.
With FS95 there was a problem caused by the landing gear dynamics.
I was having trouble getting Dr Paul Hartl's excellent F-86 to
stay on the ground after landing after converting it from FS5.1
where it performed well. In an email, Paul mentioned that he'd
found that the solution was to increase the MOI's even though
they should not be directly related. It worked but then we had
to boost aileron and elevator effectiveness to get back the flying
qualities. In FS98 I encountered worse problems with landing
gear dynamics and again the apparant fix was to boost all MOI's
by a common factor (to keep the differences of the pairs of the
proper sense). However, this new FDE by Abacus solves that problem
differently by giving us direct control of the spring load and
damping factors for the main gear and nose gear. This makes it
possible to use correct MOI's once again for a considerable improvement
in handling realism.
Contributed by Tom Goodrick.
NOTE ABOUT VMO
The VMO variable in FDE is an indicated airspeed
that pilots use as a limit during climb. You should not mistake
this for the true airspeed associated with the max Mach or MMO
because that is the default value that shows up when you first
look at the .air file using FDE.
VMO should be provided by the aircraft flight
dynamics designer. To estimate it, use the indicated airspeed
at max cruise and add 30 to 40 knots. VMO is based on the maximum
dynamic pressure the aircraft can take safely.
Submitted by Tom Goodrick.
400 - CAN YOU HELP???
The 714th Virtual Flight Test Squadron is
looking into Section 400 which seems to be a lift/drag table
which should enable the modeling of airfoils. The table seems
to be made of x,y pairs. The y-value seems to be the coefficient
of drag. It would seem that the x-value would represent the coefficient
of lift, but as CL increases so should CD. If you know what the
x-value represents, please contact Greg
Greg Pierson says that one nice thing about
this table is that you can set the stall speed. By adjusting
the maximum CL, you can adjust the stall speed and then smooth
the L/D curve.
Greg also think he's spotted Table 404 which
looks like Angle of Attack vs Coefficient of Lift; Table 430
which is drag rise due to high speed compessibility; Table 433
which high speed nose tuck.
Submitted by Greg
Pierson of the 714th Virtual Test Squadron. Greg also recommends
another nice reference called The
Incomplete Guide to Airfoil Usage.
LANDING GEAR PARAMETERS
Tom Goodrick has submitted this goodie.
The landing gear parameters in FDE allow you
to cure a bad case of jumpy gear - where the aircraft rocks or
jumps with little or no excitation. This often becomes evident
only when landing with low fuel.
- Calculate the maximum weight: W = WD + F
where WD = Dry Weight Section 1101-28
F = total fuel gallons
FWG = fuel weight per gallon (6.0 for piston, 6.6 for jet)
- Main Gear Spring Loading Factor = 2.2 * W
- Main Gear Damping Factor = 0.66 * W
- Center Gear Spring Loading = 0.6 * W
- Center Gear Damping Factor = 0.075 * W
Use these estimates to get close to the most
desireable settings and then adjust them slightly as needed.
In an aircraft landing simulation, the damping factor, which
produces load in proportion to vertical rate, is especially important
because it gives the initial touchdown load before the gear has
compressed any. A high damping factor can cause a very high initial
LIKE YOUR FEEDBACK
If you'd like contribute to our knowledge
bank, please address your feedback including suggestions, corrections
and problems to Abacus Tech Support at firstname.lastname@example.org.
Thanks for your participation.