Question about SGI flight simulator
Roy Smith
We just got a new SGI Iris, so naturally I've been playing around
with the flight simulator. I'm not a pilot, so I don't know if my
perception that it's all wierd is correct. I figured I might find some
people on rec.aviation who were pilots who had access to Irises who might
be able to tell me for sure.
The only way I've found to make the damn plane turn is to roll
towards the direction you want to go and pull back on the stick. Rudder
angle doesn't seem to make any difference at all. I was under the
impression that the way to make a turn is to roll and use the rudder, not
roll and use the elevators. I think I have a pretty good feel for where
all the lift and thrust vectors are (I'm a sailor; sailboats are just
planes on their side with two big wings in the air and two little ones in
the water).
Also, I've managed to run out of fuel in the F-15 and glide for
what seems like miles. I always thought a plane like the F-15 fell like a
rock when it lost engine power.
So, which is it? Is the flight simulator all screwy, or is my
"never having done it myself" perception of how a plane files all screwey?
--
r...@alanine.phri.nyu.edu (Roy Smith)
Public Health Research Institute
455 First Avenue, New York, NY 10016, USA
"Arcane? Did you say arcane? It wouldn't be Unix if it wasn't arcane!"
Ed Falk
> So, which is it? Is the flight simulator all screwy, or is my
>"never having done it myself" perception of how a plane files all screwey?
The flight simulator is all screwy. My theory is that they sold it
to the military, and the military made them rip out all realistic
simulation code for security reasons. Just a guess though.
Many years ago they showed it at siggraph built into an airplane
cockpit. They set the fuel to 60 seconds worth to keep the line
moving. So they had a contest to see if anybody could take off in a
150, reach some altitude, turn around and land again without running
out of fuel. I was the only one present who could do it. I did an
immelman turn on takeoff. Don't try this at home, kids.
-ed falk, sun microsystems -- sun!falk, fa...@sun.com
"What are politicians going to tell people when the
Constitution is gone and we still have a drug problem?"
-- William Simpson, A.C.L.U.
Yiannis Papelis
>with the flight simulator.
>
>towards the direction you want to go and pull back on the stick. Rudder
>angle doesn't seem to make any difference at all.
We have several Iris's (VGX and PI's) at work and I tend to agree with
you. I think the model that is used is not realistic at all. I don't have
any experience flying jets but I have flown Warriors in reality and several
other flight sims (microsoft, ACM, Next ...). The prop aircraft don't
behave anywhere near realistic.
Often you can yaw the plane and bank it and it won't show
any turning tendencies until the bank is very large. I still can't land
a single plane on the runway simply because I can't turn at low altitude ...
Oh well... I still find it exciting to fly the F-15 through the mountains
turning with bank and elevator only. It is especially impressive in our
setup where the iris is driving a projector that displays on a 8x8' screen.
BTW, after flying real airplanes I have come to the conclusion that
flying the low fidelity sims on PCs and workstations is much harder than
flying actual planes. I have never flown on an high fidelity sim though.
Anybody else has opinions on that ?
--
Yiannis E. Papelis -------- Electrical & Computer Engineering
yia...@ccad.uiowa.edu -------- University of Iowa
Gary Tarolli
> In article <1991Oct16....@phri.nyu.edu> r...@alanine.phri.nyu.edu (Roy
Smith) writes:
>
> > So, which is it? Is the flight simulator all screwy, or is my
> >"never having done it myself" perception of how a plane files all screwey?
>
> The flight simulator is all screwy. My theory is that they sold it
> to the military, and the military made them rip out all realistic
> simulation code for security reasons. Just a guess though.
>
planes have the correct weight, thrust, and wing area. The modeling of
some forces (lift, drag, air density) is fairly accurate. Some things
like rudder are not modelled at all. Note that some effects vary
with the wing shapes and angles. I believe if you roll an f-15 with
flat wings 15 degrees it wont automatically turn (a lot). Planes
with wing dihedral angles do automatically start turning and also self-level
themselves. The flight simulator assumes flat wings and doesn't model
the dihedral effect or asymetric wing lift. Hey its a game! not a product!
I do not know how an F-15 glides in real life, but I believe its modelled
pretty accurately. The glide ratio of a plane can be computed as some
simple ratio of wing area, weight and some other stuff. The C-150
drops like a rock and the B-747 glides wonderfully. This is counter-
intuitive but accurate I believe. That's why C-150 always end up landing
on highways (they can never make it to airports), and a B-767 once glided
60 miles to an airport when it ran out of fuel.
______________________________________________________________________________
_____ ______ _ _ (508)562-4800 tar...@sgi.com
/ ___ __ __ / __ __ ___ // // * (508)562-4755 (fax machine)
(____/ (_/_/ (_(_/ / (_/_/ (_(_/ (/_(/_/_ M/S DER-200
_/
Mary Shafer
In article <1991Oct17....@ccad.uiowa.edu>, yia...@ccad.uiowa.edu (Yiannis Papelis) writes:
|> Oh well... I still find it exciting to fly the F-15 through the mountains
|> turning with bank and elevator only. It is especially impressive in our
|> setup where the iris is driving a projector that displays on a 8x8' screen.
Remember that fighter pilots keep their feet on the floor. There's an
ARI to take care of the rudder.
|> BTW, after flying real airplanes I have come to the conclusion that
|> flying the low fidelity sims on PCs and workstations is much harder than
|> flying actual planes. I have never flown on an high fidelity sim though.
|> Anybody else has opinions on that ?
Having flown both low and high fidelity fighter sims, as well as fighters,
I'd certainly agree that the PC sims are extremely difficult to fly.
The high fidelity sims, with or without domes, are quite easy to fly
and do fly like the actual fighters.
>* A keyboard/mouse is a poor substitute for a stick and rudder. A mouse
> (or joystick for that matter) does not push back harder as your speed
> increases.
Remember that in fighters, there's no speed feedback. The feedback is
g. In fact, stick force per g (F_stk_lat/g and F_stk_long/g are
specified in the MILPRIME) is the parameter of interest. This is to
keep the pilot from pulling the wings off. The F-16's force stick
couldn't be scheduled with g, so they put g limiters in the control
system and the pilots hated it so much that they went to a position
stick. (The F-8 and A-7 use bobweights to get the right stick forces,
BTW.)
Just as an interesting side note, we (Dryden) got the source code for
Flight Simulator and use it for our space positioning during flights.
We coded in the terrain (the entire western US, coarsely, down to the
runways at Edwards, in detail), with all the restricted areas and
other landmarks. We generate the ground track during flight and can
also record area restrictions, etc.
--
Mary Shafer DoD #0362 NASA Ames Dryden Flight Research Facility, Edwards, CA
sha...@skipper.dfrf.nasa.gov sha...@pioneer.arc.nasa.gov
Of course I don't speak for NASA
"Turn to kill, not to engage." CDR Willie Driscoll
"Hey, Willie, how long can you tread water?" CDR Randy Cunningham
Riley Rainey
I'd have to agree that PC/workstation simulators are much harder to fly.
In my mind, several things contribute to this:
* It's difficult to render visual scenes fast enough so that the pilot
can react very smoothly on the controls.
* A keyboard/mouse is a poor substitute for a stick and rudder. A mouse
(or joystick for that matter) does not push back harder as your speed
increases.
* One does not pull-Gee's at one's desk. It is harder to understand
what a simulated plane is doing at any point in time (especially for
someone new to that particular software) when you don't feel what's
going on, as well as see.
As the author of ACM, I've probably logged more than my share of time in
front of a hurtlin' piece of silicon. I'd still rather fly a real
Cessna, though.
--
Riley Rainey Internet: ri...@mips.com
MIPS Computer Systems Phone: +1 214 770-7979
Dallas, Texas
Heiner Biesel
>In article <1991Oct17....@ccad.uiowa.edu>, yia...@ccad.uiowa.edu
(Yiannis Papelis) writes:
>|> In article <1991Oct16....@phri.nyu.edu> r...@alanine.phri.nyu.edu (Roy Smith) writes:
...[ ]...
>|> BTW, after flying real airplanes I have come to the conclusion that
>|> flying the low fidelity sims on PCs and workstations is much harder than
>|> flying actual planes. I have never flown on an high fidelity sim though.
>|> Anybody else has opinions on that ?
Having worked in the past as the project engineer responsible for the visual
systems of several different flight simulators, including the B-2, and the
F-16 at General Dynamics in Ft. Worth, I'll contribute my two cents worth.
>I'd have to agree that PC/workstation simulators are much harder to fly.
>In my mind, several things contribute to this:
>* It's difficult to render visual scenes fast enough so that the pilot
> can react very smoothly on the controls.
Precisely. That's why the big boys spend millions of dollars on the visual
system alone when they plan to do ergonomic studies, or evaluate alternative
designs. Then they spend millions more on complex visual data bases over which
to fly.
>* A keyboard/mouse is a poor substitute for a stick and rudder. A mouse
> (or joystick for that matter) does not push back harder as your speed
> increases.
The quality of the interface between the pilot and the simulation makes
all the difference. This is obvious even in simple video games.
>* One does not pull-Gee's at one's desk. It is harder to understand
> what a simulated plane is doing at any point in time (especially for
> someone new to that particular software) when you don't feel what's
> going on, as well as see.
Kinesthetic feedback, on the other hand, does not seem to be as important
in creating an effective illusion of flying as good visuals, a good model
of the dynamics of the aircraft, and good controls. Many very effective
simulators get by with just good visuals and dynamics. Occasionally, a
g-suit or g-seat is used for the hotter fighters, but motion platforms
are de rigeur primarily for commercial airline simulators, largely due to
FAA requirements to qualify for category II.
I have personally flown several different high-fidelity simulators, and
have discussed at length the characteristics of the real aircraft vs. the
simulated dynamics with the test pilots of these programs, and the concensus
is that the simulations are very close indeed. If the real aircraft shudders
and begins to stall at 110 knots IAS dirty, the same thing will happen in
the simulated aircraft.
In simulation, the first question is always "How much is good enough?",
and the second question is "What can you afford?". The best is remarkably
good, and inordinately expensive. Fortunately, the costs of simulators
continue to decline rapidly, to the point where the visual simulation
available from a computer workstation today begins to approach what only
ten years ago cost over a million dollars.
Regards,
Heiner bie...@thrall.sim.es.com
Lawrence H. Miller
>In article <1991Oct16....@phri.nyu.edu> r...@alanine.phri.nyu.edu (Roy Smith) writes:
>
>> So, which is it? Is the flight simulator all screwy, or is my
>>"never having done it myself" perception of how a plane files all screwey?
>
>The flight simulator is all screwy. My theory is that they sold it
>to the military, and the military made them rip out all realistic
>simulation code for security reasons. Just a guess though.
>
>Many years ago they showed it at siggraph built into an airplane
>cockpit.
Let's see if I can remember this. I was the one that designed
and had built the flight controls for the SGI cockpit. I did
this at ISI where I was running the Pilot's Associate program.
We built a joystick/throttle quadrant and hooked it up to the
SGI simulator. Jim Clark, founder of SGI, is a pilot, and liked
our stuff, so we'd crate up the joystick/throttle and ship it off
to them whenever they needed it for a show.
The original SGI aerodynamics were more or less based on
theory, taken from Shevell's "Theory of Flight." We found
this inadequate for our needs, and I ended up rewriting all
of the simulator code except the out-the-window stuff. I
added navigation, turbulence, clouds, highway in the sky,
completely redid the instrument panel using predictive and
directive displays, heads up displays, second terminal used
to control flight parameters, and on and on.
We simplified the aircraft handling. In the SGI version,
when you pull back on the stick, you increase the angle of
attack and that generates more lift, and so on. On our
version, we just increased the pitch, and had a formula for
relating that to rate of climb and airspeed. The flying
properties were more realistic, and the equations were a lot
simpler. But the project's purpose was investigating the
pilot computer interface.
--
Larry Miller
The Aerospace Corporation
lmi...@aero.org
(213 soon to be 310)336-5597
Richard Bartels
> We just got a new SGI Iris, so naturally I've been playing around
>with the flight simulator. I'm not a pilot, so I don't know if my
>perception that it's all wierd is correct. I figured I might find some
>people on rec.aviation who were pilots who had access to Irises who might
>be able to tell me for sure.
I griped about this a year or more ago. Your perception is correct:
the planes are wierd. It is the case (as I learned when I was
taking flying lessons) that planes are initially non-intuitive
in some of their behavior, but they don't behave the way
the ones in the flight simulator do.
> The only way I've found to make the damn plane turn is to roll
>towards the direction you want to go and pull back on the stick. Rudder
>angle doesn't seem to make any difference at all. I was under the
>impression that the way to make a turn is to roll and use the rudder, not
>roll and use the elevators. I think I have a pretty good feel for where
>all the lift and thrust vectors are (I'm a sailor; sailboats are just
>planes on their side with two big wings in the air and two little ones in
>the water).
In a plane, for steep turns, you roll, use the rudder, use the elevators,
and adjust throttle, all in concert. The plane is balanced around a
center of gravity roughly in the middle of the cabin (plus or minus,
one of the careful flight preps is to figure out where the weight
goes so that the center of gravity falls within an acceptable envelope).
In level flight, thrust, drag, lift, and gravity are all in balance about
that center. The c. o. g. and the forces are not that simple,
of course -- lift is the integral of all the differential forces
acting over the wings, and they cover quite an area, and
provide some interesting moment arms -- but in simple
flight training you don't get into the deep physics. However,
the deep physics makes for some of the non-intuitive action.
And the deep physics is what the simulator lacks. It even lacks
some of the shallow physics.
Take a mild turn. You roll slightly, say 5 degrees. The
vector integral of the differential lift components now sum to
a force that is canted 5 degrees to the side. Lift is now
slightly reduced. Moments are slightly skewed. The plane
starts to precess slowly around its vertical axis. No doubt the
vertical and horizontal stabilizer (tail and rear wings) also
do their bit in this. The net effect in much like a gradual turn
in a car. Your heading is altered at a rate of degrees per second
that depends on your forward velocity. You also loose a tiny bit of
altitude. Normally you do this only for a few degrees, so
that is not noticed. Up and down drafts account for more to pay
attention to minute by minute, so you are constantly fine tuning
altitude. You don't worry about rudder, elevators, or throttle.
Take the flight simulator. Roll 5 degrees to the right.
You will fly straight forever at a tilted setting, viewing the
horizon at a 5 degree slant. You can achieve this attitude
in a real plane only by counteracting the turning precession by
pressing "opposite rudder" slightly. Roll to the right and
press the left rudder pedal. This is known as "using crossed controls."
Take a steeper turn. You roll 15 degrees or so. Now the lift
has decreased to the cos(15) times what it was, and has become
more noticeable. You have to draw back on the control column
slightly to oomph the lift up a bit. Now your speed will drop
slightly in contrast to your altitude. Again, not really worth
the notice, if you don't keep it up too long.
On the other hand, the lift on that long boom out your rear
(a.k.a. the vertical and horizontal stabilizers) has also
dropped by cos(15) whatever, and the plane starts to skid sideways.
You compensate by adding a bit of camber to the vertical stabilizer
profile to oomph its lift up a bit, too. In direct terms, you
press on the rudder pedal "in the turn direction". Roll right,
press right pedal. There is a neat gizmo that helps you coordinate
what has to be done. It is called (guess what) a "turn coordinator".
It is really two things, a canted gyro and a little rubber ball
in a U-shaped tube of fluid. The gyro tries to precess, and measures
the rate of turn. The ball slides to the right or left and gives
an indication of the slip. You try to keep a constant turn rate and
hold the ball at the bottom of the U. You even try to do it while the
instructor is making disparaging comments.
Take the flight simulator. Roll 15 degrees or so.
Same old story -- you stare at a slanted horizon and go straight on.
You don't turn unless you "climb into the roll," which is not what
the real thing is doing.
Now go for the big time. You roll to 45 degrees or more.
The lift goes down to half or below and the rate of precession
about the vertical axis goes up dramatically. You have the
impression that you are turning on a dime, and you are beginning to
feel significant G forces. You remind yourself of where the barf bag
is stashed. You have to compensate by pressing on the
rudder forcefully, pulling back firmly, and upping the throttle
to maintain velocity. It becomes a delicate balancing act.
Too little back elevator, and you drop. Enough back elevator, but
insufficient extra throttle, and you slow down and drop.
In either case, trying to save the day by pulling back hard once
the dropping sets in, and you simply tighten the spiral you have
begun. Your speed builds up dramatically, but it is directed
downward. Since planes fall apart if they go too fast, and
damage themselves if they do things quickly at somewhat lesser
speeds, you learn to apologize to the instructor, kill the
throttle, turn up the carb heat, and ease gently out of the dive --
not necessarily in that order.
Be successful with elevator and throttle, and you can still make
a mess with the rudders, slewing around the turn,
but this is less serious.
Take the flight simulator -- please.
As for other gripes, the simulator assumes that a stall spells
disaster, producing a chaotic fall from the sky.
What a fighter plane does, I have no idea, but private craft
are built to be very stable. The nose drops, the plane does
a controlled belly flop for a hundred feet or so, and usually
picks up enough speed to wallow on. You, of course, don't let
it muddle on like that. Push the yoke sharply to drop the nose
with emphasis and push the throttle, and you can be on your way
with minimal altitude loss. Near stall speed, by the way,
the plane reacts very sluggishly to controls, something the simulator
gives no hint of. Near the stall speed, prudence dictates that
you keep the wings level, since rolling is not very dependable,
and you hold the heading with the rudders.
Here, again, the simulator sucks. Try the rudders in level
flight and see if your heading changes. Just as in roll, yaw
will change the attitude of the simulator plane, but not change
its flight path.
Finally, suppose you do stall into the catastrophic. Congratulations,
you are now in a spin. This is a suprisingly stable configuration for
a plane to be in, and not the chaotic state the sumulator makes it
out to be. One wing on a real plane is stalled (providing drag but no lift)
and the other wing is still providing lift. This imbalance
corkscrews the plane about some axis I still haven't figured out,
but the view out the window is spectacular. Someone has
put the world on a turntable at 33.333 rpm. and it is getting
bigger. Unless you do something, you will stay in this stable
configuration, losing altitude, looking at the turntable, until you
plow in. On a really stable plane it is hard to get here in the
first place (at the point of stall, yank back on the yoke and slam
hard on one rudder pedal -- the aeronautical equivalant of hitting
the plane over the head with a sledge hammer). On a stable plane,
getting out requires decisive rudder press opposite the sense of
rotation, sharp forward on the yoke, and you are left in a dive that
you pull gingerly out of.
The instructor could always get out of a spin before a complete revolution
was done with, and at a hundred feet or less of altitude loss.
If you feel a spin coming on, yoke and rudder can prevent the state
from ever ocurring -- much like the automatic reaction most of us
have on icy roads that kills the start of a skid.
Nothing like this ever happens on the sgi flight simulator.
> Also, I've managed to run out of fuel in the F-15 and glide for
>what seems like miles. I always thought a plane like the F-15 fell like a
>rock when it lost engine power.
Beats me, having only flown Cessnas and Pipers. First thing on these,
if you run out of gas, the Department of Transport cites you for
violating the rule of "having sufficient fuel to proceed to your
destination and thence 3/4 hour in reserve."
But the instructors revel in pulling the throttle off at the most
unexpected moments. Like a bunny you: (1) put carb heat on (2) trim
to best glide velocity (3) scope out a landing field and a base point
(4) run a check on motor and try to restart (5) switch to emergency
frequency and report (6) set up for your final approach and
(7) glide in. At 3000+ above ground and in practice you have ample
time to do all that. Cessna 172's glide at 65 knots and
can stay up for a bit. Landing with no power, you can afford to
be slower than that on the final few yards. Full flap stall speed
is 33 knots.
> So, which is it? Is the flight simulator all screwy, or is my
>"never having done it myself" perception of how a plane files all screwey?
The simulator is all screwey. It doesn't have to be. You can buy
PC based flight simulators that supposedly have reasonable flight
characteristics -- they are advertized in flight magazines as
a reasonable, informal way to keep up on instrument skills (though not
legally acceptable as flight time).
-Richard
William LeFebvre
|> I'd have to agree that PC/workstation simulators are much harder to fly.
As do I.
|> In my mind, several things contribute to this:
[ lists three excellent reasons ]
Other reasons:
* the view on a computer is two-dimensional. There is no depth
perception, making it very difficult to judge distances.
* the view is very limited. When I fly a pattern for real, I am
constantly looking ahead, at 90 degrees, at 135 degrees (over
my shoulder) to judge distance, glideslope and thus throttle and
pitch. Doing this on a computer requires much more than just
turning your head.
I've found that succesfully flying a pattern all the way to a landing
on a computer simulator is almost impossible. Straight in approaches
are actually easier, even without the help of an ILS or a VASI,
probably because it is easier to get "the big picture" (see point 3
above).
William LeFebvre
Computing Facilities Manager and Analyst
Department of Electrical Engineering and Computer Science
Northwestern University
<ph...@eecs.nwu.edu>
Bruce Hill
>
> We just got a new SGI Iris, so naturally I've been playing around
> with the flight simulator. I'm not a pilot, so I don't know if my
> perception that it's all wierd is correct. I figured I might find some
> people on rec.aviation who were pilots who had access to Irises who might
> be able to tell me for sure.
About a year ago we had an Iris machine and I too played around with the
simulator a bit. Great graphics, but the aircraft are really hard to
fly with any precision.
>
> The only way I've found to make the damn plane turn is to roll
> towards the direction you want to go and pull back on the stick.
Using the mouse for stick control takes an awful lot of effort.
It was very easy to over control the airplane.
I found that many times I could not get the aircraft to turn at all.
Rudder, ailerons and elevators don't work very well together. Individually
they sort of work.
> Also, I've managed to run out of fuel in the F-15 and glide for
> what seems like miles. I always thought a plane like the F-15 fell like a
> rock when it lost engine power.
I imagine that it does. How high up were you?
Have you tried exploring the service ceilings? I was able to get the
F-15 beyond 70,000'. Can the real F-15 do this?
When you start the simulator it puts you on what appears to be a runway.
After several near misses on take-off with a hangar at the end of the runway,
did I realize that it starts you on a taxiway. The toggle between tower-view
and pilot-view can allow you to watch your takeoff from the tower.
Kind of like an IFR takeoff...
The best fun I had with the simulator was doing aerobatics in the B-747.
One thing I wish they had done different is the layout of the instruments.
It has too many sliding-bar/numerical type indicators. It would be alot
nicer if they had the normal style indicators in a 6-pack instrument cluster.
> So, which is it? Is the flight simulator all screwy, or is my
> "never having done it myself" perception of how a plane files all screwey?
The simulator is screwy. After all, it is only a demo, and the primary
thing that is being demoed is the graphic capability.
--
Bruce T. Hill Danford Corp. voice: (213) 514-9334
Project Manager 350 W. 5th St. FAX: (213) 831-0454
uunet!dannet!bruce San Pedro, CA 90731 USA
Thaddeus Beier
financially-induced break from my flight instruction. The first
flight back in a real plane was pretty exciting. I was a little high
over the threshold, so I just dived for the ground (which is what
you have to do on the SGI simulator...) My instructor (for the first
time ever) grabbed the controls and said "Don't ever do that!"
I knew it was wrong, but you do get muscles remember what you
have done thousands of times in the flight simulator, and it is hard
to do what you know is right when you are really in the air. I forced
myself to stop playing at work after that ;) yeah, right.
--
Thad Beier What is good, and what is not good, need we ask anybody
to tell us these things?
scru...@ac.dal.ca
> simulator a bit. Great graphics, but the aircraft are really hard to
> fly with any precision.
Horse puckey. :-) Practice makes perfect.
> Have you tried exploring the service ceilings? I was able to get the
> F-15 beyond 70,000'. Can the real F-15 do this?
I've had the 14,15 and perhaps 16 up to 100,000 feet at which
point the program burbs. Doesn't know how to render space?
I've also had the 14 up to 7000 MPH indicated but it feels more
like 100.
One thing people seem to miss out on is that the tower view is
a great simulation of a radio controlled airplane and except for
the small rudder problem the planes react a lot like R/C ones,
except on landing. You might bounce an R/C plane 30 scale feet
and get away with it but coming down that fast in dog ends in the
obligitory explosion; which is real cheap compared to R/C.
--
-Jeff scru...@ac.dal.ca Dalhousie University Oceanography Dept.
Mike York
>intuitive but accurate I believe. That's why C-150 always end up landing
>on highways (they can never make it to airports), and a B-767 once glided
>60 miles to an airport when it ran out of fuel.
As my memory serves me, it was not an airport where the 767 landed, but an
old glider landing strip that was converted for use as a drag strip. One
of the crew members was an ex CAF (or perhaps RAF) pilot that had used the
strip many years before and luckily remembered its location.
--
Mike York |
Boeing Computer Services | Support your local nanobrewer --
(206) 865-6577 | No twist-off bottles.
zom...@voodoo.boeing.com |
Robert I. Plested 76809
You're right, the SGI flight sim is a poor model of a real aircraft. If
you're really into it, I like the Microsoft Flight Sim, I think they're
up to version 4 by now.
Bob
Gary Tarolli
> In a plane, for steep turns, you roll, use the rudder, use the elevators,
> and adjust throttle, all in concert. The plane is balanced around a
> center of gravity roughly in the middle of the cabin (plus or minus,
> one of the careful flight preps is to figure out where the weight
> goes so that the center of gravity falls within an acceptable envelope).
> In level flight, thrust, drag, lift, and gravity are all in balance about
> that center. The c. o. g. and the forces are not that simple,
> of course -- lift is the integral of all the differential forces
> acting over the wings, and they cover quite an area, and
> provide some interesting moment arms -- but in simple
> flight training you don't get into the deep physics. However,
> the deep physics makes for some of the non-intuitive action.
> And the deep physics is what the simulator lacks. It even lacks
> some of the shallow physics.
c.o.g., moments, rudder are not simulated
>
> Take a steeper turn. You roll 15 degrees or so. Now the lift
> has decreased to the cos(15) times what it was, and has become
> more noticeable. You have to draw back on the control column
depends on your frame of reference. vertical lift has decreased,
but lift in the "up" direction as viewed from the pilot's seat
has not changed significantly. Flight accurately models this,
however it doesn't do integrals. Lift is just a function of
airspeed and angle of attack and wing area.
> slightly to oomph the lift up a bit. Now your speed will drop
> slightly in contrast to your altitude. Again, not really worth
> the notice, if you don't keep it up too long.
> On the other hand, the lift on that long boom out your rear
> (a.k.a. the vertical and horizontal stabilizers) has also
> dropped by cos(15) whatever, and the plane starts to skid sideways.
> You compensate by adding a bit of camber to the vertical stabilizer
> profile to oomph its lift up a bit, too. In direct terms, you
> press on the rudder pedal "in the turn direction". Roll right,
> press right pedal. There is a neat gizmo that helps you coordinate
> what has to be done. It is called (guess what) a "turn coordinator".
> It is really two things, a canted gyro and a little rubber ball
> in a U-shaped tube of fluid. The gyro tries to precess, and measures
> the rate of turn. The ball slides to the right or left and gives
> an indication of the slip. You try to keep a constant turn rate and
> hold the ball at the bottom of the U. You even try to do it while the
> instructor is making disparaging comments.
>
> Take the flight simulator. Roll 15 degrees or so.
> Same old story -- you stare at a slanted horizon and go straight on.
> You don't turn unless you "climb into the roll," which is not what
> the real thing is doing.
>
sorry, but this is wrong. I just ran flight on my Indigo, and for
both the F-15 and the C-150, when I banked 15 degrees and continued
to fly level (required a little pull on the stick) they both
turned nice and slowly. If you use 'v' (autopilot) the computer
will keep your speed and rate of climb constant (if it can). This makes
it easy to test. However, on my PI, your complaint is valid. The
difference between the two versions is a bug fix (thanks rob) for
floating point accuracy in the heading angle computation. So
the flight equations are not as bad as you think.
>
> As for other gripes, the simulator assumes that a stall spells
> disaster, producing a chaotic fall from the sky.
> What a fighter plane does, I have no idea, but private craft
> are built to be very stable. The nose drops, the plane does
exactly - i have no idea either...!!!
I'de love to hear your comments on Atari's pole position, or
Nintendo's Rad Racer. Flight is a game. I purposely made stalls
randomly chaotic for more "game enjoyment". Especially during
dogfights when you have someone on your tail and you make sharp
turns to try to lose him, being carefull not to stall and then
scream in ecstacy as he violently stalls his plane and goes flying
into a wild spin (and hopefully the ground).
>
> Finally, suppose you do stall into the catastrophic. Congratulations,
> you are now in a spin. This is a suprisingly stable configuration for
> a plane to be in, and not the chaotic state the sumulator makes it
.
.
.
> The instructor could always get out of a spin before a complete revolution
> was done with, and at a hundred feet or less of altitude loss.
> If you feel a spin coming on, yoke and rudder can prevent the state
> from ever ocurring -- much like the automatic reaction most of us
> have on icy roads that kills the start of a skid.
> Nothing like this ever happens on the sgi flight simulator.
>
the simulator also doesn't kill you when you crash, but i'm thinking
of adding that as an enhancement :-)
>
> The simulator is all screwey. It doesn't have to be. You can buy
> PC based flight simulators that supposedly have reasonable flight
> characteristics -- they are advertized in flight magazines as
> a reasonable, informal way to keep up on instrument skills (though not
> legally acceptable as flight time).
>
you can also buy real simulators for SGI machines from our customers.
the key word is "buy", flight is free - its a demo.
P.S. I have heard a few real commericial pilots comment about flight,
and I can't remember exactly what they said, except that I was
surprised because they were happier with the flight characteristics
than I was (maybe they were just being polite!).
--------------------
Gary Tarolli
ka...@sima.sintef.no
favorite game can cause...
Me neither (know how to fly jets), but I do know several props quite
a bit... CE172, CE152 (aerobat), Pipers and Beeches... The simulation of
the performance of the C-150 is obviously not accurate, however I LOVE IT.
And, apart from having my VISA and Amex cards ready whenever anyone
can show me a single engine prop going 350+ knots, don t fly the
C-150 in "flight", but spend more time "dog"ing around with the F14-D.
I d like to think that landing an aircraft in the simulator at high
scores (90-99) somehow reflect a nice approach and general feeling of
the thrill of flying, however, NOT managing to land the simulator
would not automatically outrule that you still might be able to pass
your next bi-annual... OK guys, it is a game - but why shouldn t we
allow ourselves to have some 100% SAFE fun every now and then ?
Karl (NPPL A6282)
Sean Schur
about the SGI flight simulator. I posted this question about a month
ago and didn't get 1 response.
How do you activate the SAM's in the SGI flight simulator. I am running
version 3.3.2 version of the software on a 4D/25. I can turn on the
wireframe domes to show where the SAM missile firing ranges are (I
forget which key it is, but it's in the man pages). However, the
doc just says that they should be on in this mode, but I have never
seen a SAM fire. Is there another key to activate the SAM's? Has anyone
ever had a SAM fire at them?
=======================================================================
Sean Schur USENET: sc...@isi.edu
Assistant Director SGI/Amiga/Media Lab Compuserve: 70731,1102
Character Animation Department
California Institute of the Arts
=======================================================================
Gary Tarolli
>
> How do you activate the SAM's in the SGI flight simulator. I am running
> version 3.3.2 version of the software on a 4D/25. I can turn on the
> wireframe domes to show where the SAM missile firing ranges are (I
> forget which key it is, but it's in the man pages). However, the
> doc just says that they should be on in this mode, but I have never
> seen a SAM fire. Is there another key to activate the SAM's? Has anyone
> ever had a SAM fire at them?
>
Although you could turn on the SAM threat cones ('T'), there never was
any SAM firing code in the game. Note I said "was". Now there is.
In the next SGI release, 4.0.1, the PI and Indigo machines will have
SAM missiles. They are more deadly than sidewinders and only chase
you (even in a dogfight). Their main purpose is for fun, when no
one else wants to play. A few more details - if you already have a missile
in flight, it gets destroyed and the SAM fires. SAMs blow themselves
up when they run out of fuel (so that another can be immediately fired).
You cannot fire when a SAM is in flight (the existing code only handles one
missile - it obviously is in need of some enhancements).
And before someone complains about how inaccurately sidewinders and SAMs
are modelled, let me offer a pre-emptive strike. WHO CARES! It wouldn't
be much fun in a dogfight with Pheonix missiles that are 99% deadly.
The sidewinders and SAMs were adjusted to offer you some small chance
(actually decent if you fly fast and furious) of hope. Its tuned to
be an interesting game. Life is sometimes dull and a bitch. Flight
is recreation.
--------------------
Gary Tarolli
Mike Fitzpatrick
The only unrealistic part about the flight simulator that disturbs
me is that the C-150 has a 20mm cannon on it! Oh well, that just makes it
more fun to buzz the tower, and dog becomes a completely new game. Looking
forward to the new version....
Bruno Pape
>
>The only unrealistic part about the flight simulator that disturbs
>me is that the C-150 has a 20mm cannon on it!
>
Cessna dosn't offer a 20mm cannon as an option?
Maybe it was just my flight instructor making those rat-tat-tat sounds.
Thomas Benedict
First off, we're running 3.3.2 on a PI, so if this is already on the
one running on the VGX systems, bear with me.
I just went for my first flight in a sailplane. It was a real blast.
While we were up there was a guy in an aerobatic glider doing stunts.
I was watching him go up thermals below us and do stunts until he had
to climb again. Adding a glider to flight would be really nice,
though I can see how adding thermals might be a pain if they aren't
already there. Gliders are simply beatiful planes to fly.
The other thing I was wondering is if there was a version of flight or
an option on the current one to interface to the dial box. I wasn't
thinking of using the box so much as using the interface to construct
a set of flight controls. It would also be nice if the flight program
could write out one of the serial ports as well. Compumotor makes a
nice stepper motor controller that can be run via a serial connection.
Adding some feedback to the flight controls should improve the feel of
the simulation.
Awright, so I'm talking about a bunch of equipment for something I
only do after hours, and don't do very seriously, but it would be
nifty to have the option.
Tom Benedict
bene...@chaos.utexas.edu
Center for Nonlinear Dynamics
Nathan F. Janette
It would be nice if the cannon fired in bursts of 50 rounds or so, which would
be more like the "real thing", and if there was an appropriate ammount of ammo
for different aircraft (ie F-15 = 950 rounds, F-16 = 500 rounds, etc.)
Since we're discussing cannons, how about adding an A-10 II ground attack plane
with it's 30mm cannon. Of course, this ties into my idea that there should be
another game integrated into dog called "tank". Imagine playing with combined
air and ground operations! Any reason this couldn'd be done? It seems like
one would just cross-reference "arena" and "dog" in a programming-kind-of-way.
--
Nathan Janette
Dept MB&B
Yale University
nat...@jacobi.csb.yale.edu (NeXT)
douglas.e.drummond
Distribution: world
Organization: AT&T Bell Laboratories
Keywords:
In article <1991Oct17....@ccad.uiowa.edu> yia...@ccad.uiowa.edu (Yiannis Papelis) writes:
>In article <1991Oct16....@phri.nyu.edu> r...@alanine.phri.nyu.edu (Roy Smith) writes:
>>
{much discussion deleted}>
>BTW, after flying real airplanes I have come to the conclusion that
>flying the low fidelity sims on PCs and workstations is much harder than
>flying actual planes. I have never flown on an high fidelity sim though.
>Anybody else has opinions on that ?
>--
>Yiannis E. Papelis -------- Electrical & Computer Engineering
>yia...@ccad.uiowa.edu -------- University of Iowa
Hi -- I am a 280 hour VFR Cessna 172 pilot, computer consultant by trade,
and conseseur of simulators.
Most commercial simulatore *suck* (sorry). Chuck Yeagers is o.k. but
Rev 2.0 requires at least a 286 PC to do it justice, and some of the
parameters are WRONG, especially for the space shuttle.
IMHO, a *decent* simulator is harder to fly than a real, full scale airplane.
(Note every word in the previous sentence is significant.)
I wrote a simulator for a PDP-11/45 with Vector General graphics terminals
that was approximatly equivalent to a "software" Frasca. Various friends
"flew" it who were radio control pilots or who flew with me from the
Right Seat in the C-172. The general opinion was that it was harder than
the airplane -- about as hard as R/C airplanes.
This really was the good news because I got about 100 free simulator hours
this way. Yes, I know I can't log them but the practice in instrument scan,
etc. counts. It was helpfull that last time I did any serious instrument
training, which was about 6 years ago.
Note that this simulator used very simple equations for square-law drag and
yaw as a sin/tan function of roll angle.
I am very interested in simulators that use very simple graphics and
performance equations.
Douglas E. Drummond wb9idj\@vpnet.chi.il.us (permanent email address)
This was posted from Bell Labs
-end transmission-
Steven Philipson
rhba...@watcgl.waterloo.edu (Richard Bartels) compares flying the SGI
"Flight" simulator to flying an aircraft.
Richard, you're right that the program has unrealistic dynamics, but
some of your ideas about how to control an airplane are wrong.
> In a plane, for steep turns, you roll, use the rudder, use the elevators,
> and adjust throttle, all in concert.
Throttle is rarely used in turning. There is an increase in drag
during a turn, but usually the throttle is left at the cruise setting
and the small decrease in airspeed is tolerated. Turns with banks in
excess of 45 degrees will require a power increase if the airplane is
being flown at low speed. Some low-performance aircraft require full
throttle for banks of 60 degrees and will lose altitude even at full
throttle in steeper banks. This also depends on altitude which effects
power required and power available.
> Take a steeper turn. You roll 15 degrees or so. Now the lift
> has decreased to the cos(15) times what it was, [...]
The vertical component of lift decreases, but the lift of the
wing (and tail) does not change.
> [...] You have to draw back on the control column
> slightly to oomph the lift up a bit.
The objective is to increase total lift such that the vertical
component of lift is equal to aircraft weight.
> On the other hand, the lift on that long boom out your rear
> (a.k.a. the vertical and horizontal stabilizers) has also
> dropped by cos(15) whatever, and the plane starts to skid sideways.
No. If you do not add up elevator the aircraft will descend, but
this will not effect turn coordination.
> You compensate by adding a bit of camber to the vertical stabilizer
> profile to oomph its lift up a bit, too. In direct terms, you
> press on the rudder pedal "in the turn direction".
Rudder is used to correct for adverse yaw generated by aileron
deflection. If aileron deflection is small but held as the airplane
rolls, the ball will remain centered (or close to it) even as the
airplane reaches large bank angles. Large aileron deflections will
generate adverse yaw (yaw away from the direction of the turn) which
must be countered with rudder into the turn. The rudder input is
removed when the aileron input is removed.
Aircraft with long wings (e.g. sailplanes) will begin to experience
adverse yaw from wingtip speed/drag differences in the turn at large
bank angles. This requires applying and holding rudder into the turn.
There is also an overbanking tendency due to these wingtip speed
differences. This is countered with aileron to the outside of the turn.
Sailplanes in steep banks may require large aileron deflections away from
the turn to avoid overbanking with simultaneous large rudder deflection
into the turn to counter adverse yaw.
> You roll to 45 degrees or more. The lift goes down to half or below [...]
The vertical component of lift at 45 degrees bank is 0.71 of that of
wings-level flight if angle of attack and airspeed are not changed.
> [...] You have the impression that you are turning on a dime,
Not in most aircraft. 45 degree bank turn radius is still fairly
large.
> and you are beginning to feel significant G forces.
45 degrees produces 1.4 g, which most people will notice but not
find uncomfortable.
> [...] You have to compensate by pressing on the rudder forcefully,
No. Rudder deflection depends on aileron deflection and not on
bank angle. Most light airplanes require very little rudder deflection
in a 45 degree, constant bank turn.
> Enough back elevator, but
> insufficient extra throttle, and you slow down and drop.
The aircraft will only descend if the airspeed decreases to a point
lower than the minimum power required airspeed (corrected for load
factor). This speed is much slower than typical cruise airspeeds.
One can run into this problem at high altitude where there is less
excess power available though.
> As for other gripes, the simulator assumes that a stall spells
> disaster, producing a chaotic fall from the sky.
This is indeed an unrealistic aspect of this program.
> What a fighter plane does, I have no idea, but private craft
> are built to be very stable.
There are numerous models (and individual aircraft) that are not
particularly well behaved at stall. Some will stall straight ahead,
drop the nose and increase airspeed, others will "mush" and stay in
that attitude, others will always drop the same wing, and some will
drop a different wing each time. Some provide lots of aerodynamic
warning, and others provide almost none.
> Push the yoke sharply to drop the nose
> with emphasis and push the throttle, and you can be on your way
> with minimal altitude loss.
Most Cessnas and Pipers only require that back pressure on the
elevator be relaxed. Pushing the yoke forward will produce a rapid
pitch down and excessive altitude loss.
> [...] Near the stall speed, prudence dictates that
> you keep the wings level, since rolling is not very dependable,
Stalls can be and are demonstrated in banks. These are required
maneuvers for the private pilot rating.
> Finally, suppose you do stall into the catastrophic. Congratulations,
> you are now in a spin. This is a suprisingly stable configuration for
> a plane to be in, and not the chaotic state the sumulator makes it
> out to be.
This depends on the airplane. Some have oscillatory spin states
and will change from one spin mode to another.
> > Also, I've managed to run out of fuel in the F-15 and glide for
> >what seems like miles. I always thought a plane like the F-15 fell like a
> >rock when it lost engine power.
An F-15 will glide for miles, but probably not as far as they will
in Flight. They'll also "drop like a rock" in reality -- the glide ratio
is fair, but the best glide speed is high. Thus they come down very
fast even though the glide angle isn't too bad.
> [In the 172] Full flap stall speed is 33 knots.
Flaps down stall speed in the 172 is 43 knots calibrated even in the
oldest/lightest aircraft. *Indicated* stall speed is lower, but
reflects airspeed system errors at high angle of attack.
> > So, which is it? Is the flight simulator all screwy, or is my
> >"never having done it myself" perception of how a plane files all screwey?
> The simulator is all screwey.
Yes and no. There are parts of the flight envelope that are handled
poorly, but some aspects are fair. High speed flight in the fighter
aircraft isn't too bad. The actual aircraft are be flown in a manner
that some call "bank and yank", i.e., aileron for bank angle and
elevator for pitch. The pilot's feet can remain on the floor as there
are yaw-dampers that keep the aircraft in coordinated flight. Low bank
angles should provide slow turn rates, and the simulator program didn't
do this. I believe that some recent versions have this fixed.
Steve
(the certified flying fanatic)
ste...@decwrl.dec.com
Jon Thackray
Path: harlqn!qmw-cs!qmw-dcs!icdoc!uknet!mcsun!uunet!decwrl!pa.dec.com!emu.pa.dec.com!stevenp
From: ste...@decwrl.pa.dec.com (Steven Philipson)
Newsgroups: comp.sys.sgi,rec.aviation
Date: 5 Nov 91 04:19:02 GMT
References: <1991Oct17....@watcgl.waterloo.edu> <1991Oct16....@phri.nyu.edu>
Sender: ne...@PA.dec.com (News)
Reply-To: ste...@decwrl.pa.dec.com (Steven Philipson)
Organization: DEC Palo Alto
Lines: 158
Xref: harlqn comp.sys.sgi:5163 rec.aviation:36
In article <1991Oct17....@watcgl.waterloo.edu>,
rhba...@watcgl.waterloo.edu (Richard Bartels) compares flying the SGI
"Flight" simulator to flying an aircraft.
Richard, you're right that the program has unrealistic dynamics, but
some of your ideas about how to control an airplane are wrong.
> In a plane, for steep turns, you roll, use the rudder, use the elevators,
> and adjust throttle, all in concert.
Throttle is rarely used in turning. There is an increase in drag
during a turn, but usually the throttle is left at the cruise setting
and the small decrease in airspeed is tolerated. Turns with banks in
excess of 45 degrees will require a power increase if the airplane is
being flown at low speed. Some low-performance aircraft require full
throttle for banks of 60 degrees and will lose altitude even at full
throttle in steeper banks. This also depends on altitude which effects
power required and power available.
I think most training organisations would disagree with you. Most
teach the coordinated use of the three primary controls and power for
turns with bank angles greater than 30 degrees (ie 'steep turns'). For
some larger/more powerful aircraft you can get away without use of
power, but you are probably into flying large singles or twins before
this is 'correct' procedure.
> Take a steeper turn. You roll 15 degrees or so. Now the lift
> has decreased to the cos(15) times what it was, [...]
The vertical component of lift decreases, but the lift of the
wing (and tail) does not change.
You're being picky. The intended meaning of the original was quite
clear.
> [...] You have to draw back on the control column
> slightly to oomph the lift up a bit.
The objective is to increase total lift such that the vertical
component of lift is equal to aircraft weight.
Same again
> On the other hand, the lift on that long boom out your rear
> (a.k.a. the vertical and horizontal stabilizers) has also
> dropped by cos(15) whatever, and the plane starts to skid sideways.
No. If you do not add up elevator the aircraft will descend, but
this will not effect turn coordination.
> You compensate by adding a bit of camber to the vertical stabilizer
> profile to oomph its lift up a bit, too. In direct terms, you
> press on the rudder pedal "in the turn direction".
Rudder is used to correct for adverse yaw generated by aileron
deflection. If aileron deflection is small but held as the airplane
rolls, the ball will remain centered (or close to it) even as the
airplane reaches large bank angles. Large aileron deflections will
generate adverse yaw (yaw away from the direction of the turn) which
must be countered with rudder into the turn. The rudder input is
removed when the aileron input is removed.
I think you should review the use of rudder for balance. Whilst not as
much is required in powered aircraft as in gliders where there is no
slipstream, all turning requires some use of rudder away from the
balanced straight and level configuration. This is not simply to
correct for adverse yaw, which many types correct for you by means of
frieze and differential aileron movement, but to cancel the inevitable
sideslip which will occur whenever the aircraft is banked. This is of
course to some extent removed by the secondary effect of bank caused
by the vertical fin, but not completely. If you don't believe me, try
flying a C172 (I assume you can do this) with full flap at ISA 50 mph
and 30 degrees of bank. Not only will you require plenty more power,
but vast quantities of the relevant rudder.
[stuff deleted]
> and you are beginning to feel significant G forces.
45 degrees produces 1.4 g, which most people will notice but not
find uncomfortable.
Many non-pilots find the G they experience in a 45 degree steep turn
uncomfortable, and some even find it disturbing.
> [...] You have to compensate by pressing on the rudder forcefully,
No. Rudder deflection depends on aileron deflection and not on
bank angle. Most light airplanes require very little rudder deflection
in a 45 degree, constant bank turn.
Incorrect, as above
[more stuff deleted]
Suggest you review steep turns in a light aircraft such as a C150,
and slow flight and use of rudder.
Steve
(the certified flying fanatic)
ste...@decwrl.dec.com
--
Jon
martin%bo...@edwards-tems.af.mil
> <1991Oct16....@phri.nyu.edu> <1991Nov5.0...@PA.dec.com>
>
> > Take a steeper turn. You roll 15 degrees or so. Now the lift
> > has decreased to the cos(15) times what it was, [...]
>
> The vertical component of lift decreases, but the lift of the
> wing (and tail) does not change.
>
> You're being picky. The intended meaning of the original was quite
> clear.
This is not being picky, the difference is very important to understanding
the dynamics of flight. It is important to know that during turning flight
there is a vertical component of lift (which hopefully equals weight most of
the time) and a horizontal componet of lift (which causes the airplane to
turn). Without the rotation of the lift caused by banking the airplane and the
resulting horizontal componet of lift, you have no turn.
>
> > [...] You have to draw back on the control column
> > slightly to oomph the lift up a bit.
>
> The objective is to increase total lift such that the vertical
> component of lift is equal to aircraft weight.
>
> Same again
Steve is right again.
Steve is still right. Airplanes that use spoilers for roll control exhibit
proverse yaw. That is, the pilot must apply rudder out of the turn to remain
coordinated. By using both spoilers and ailerons, as some airplanes do, it is
possible to get a configuration which exhibits neither adverse nor proverse yaw
and thus requires no rudder for turn coordination. Many complex aircraft use
an aileron-rudder interconnect which automatically puts in rudder proportional
to aileron (either into or out of the turn as required by the configuration) in
order to make the pilot think that no rudder input is required for coordinated
turning. Sideslip is not an inevitable consequence of banking the airplane, it
results from the fact that we don't know how to roll an airplane without
increasing the lift, and consequently the drag due to lift, on one side and
reducing them on the other. When you have more drag on one side of the
airplane than the other, a yawing moment is created.
All of the above assumes up and away flight, slow flight is more complex.
At slow speeds and high power settings you have three factors trying to yaw the
airplane to the left (assuming a conventional general aviation airplane):
propeller slipstream impingement on the rudder, higher local angle of attack on
the descending blade of the propeller than on the ascending, and gyroscopic
moments produced by the engine and propeller. Consequently, considerable right
rudder must be held while flying at low speeds with high RPMs and additional
right rudder will be required when initiating a turn to the right, less will
be required when initiating a turn to the left. The relative amount of rudder
required to initiate a turn during slow flight will be greater than in cruising
flight because more aileron deflection is required to produce the desired roll
rate and because a given amount of down aileron will produce more drag there
(and a given amount of up aileron will produce a larger decrease in drag on the
other side, thus providing the yawing moment).
> [stuff deleted]
>
> > and you are beginning to feel significant G forces.
>
> 45 degrees produces 1.4 g, which most people will notice but not
> find uncomfortable.
>
> Many non-pilots find the G they experience in a 45 degree steep turn
> uncomfortable, and some even find it disturbing.
>
> > [...] You have to compensate by pressing on the rudder forcefully,
>
> No. Rudder deflection depends on aileron deflection and not on
> bank angle. Most light airplanes require very little rudder deflection
> in a 45 degree, constant bank turn.
>
> Incorrect, as above
>
Nope, Correct as above.
> [more stuff deleted]
>
> Suggest you review steep turns in a light aircraft such as a C150,
> and slow flight and use of rudder.
>
> Steve
> (the certified flying fanatic)
> ste...@decwrl.dec.com
> --
>
> Jon
--
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Gary S. Martin ! (805)277-4509 DSN 527-4509
6510th Test Wing/TSWS ! Mar...@Edwards-TEMS.af.mil
Edwards AFB, CA 93523-5000 !
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Steve Lehar
> By using both spoilers and ailerons, as some airplanes do, it is
> possible to get a configuration which exhibits neither adverse nor
> proverse yaw and thus requires no rudder for turn coordination. Many
> complex aircraft use an aileron-rudder interconnect which
> make the pilot think that no rudder input is required for coordinated
> turning.
All this is correct, although of course it is more complicated than
this (what else is new?) because since the adverse yaw is caused by
unequal lift, it is therefore also modulated by the total amount of
lift. So, if you are pulling a couple of G's when you throw in your
aeleron deflection then you will need MORE rudder for the same amount
of aeleron because there is more total lift being generated.
Conversely, if you "push over the top" and get light in your seat
while slamming the aeleron to the stop, you will need hardly any
rudder. If you actually go to zero G you will need no rudder at all
(or for instance if you are going straight up, like in the rolling
part of a cloverleaf), and if you are pulling NEGATIVE G's you
actually need rudder in the OPPOSITE direction to your aeleron! All
this makes it difficult to design an automatic system to compensate in
all situations.
To "put you in the pilot's seat", here is how it looks to do a level
turn while flying upside down. Say you are flying along upside down
towards point A on the ground (above you) and would like to make a
turn towards B. (Note how the nose is pointed down (up) away from the
ground in level flight to compensate for wing incidence angle)
GROUND A B
- - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - - - -
---------------------------------------------------------------------
HORIZON
SKY + <-- the direction the plane is pointing
What you have to do is roll the aeleron to the LEFT while applying
RIGHT rudder until you have the bank angle established.
SKY \ GROUND
\
\
\
\
\
\ A
\
\
\
\
+ \
\
\
\ B
\
\
With the proper bank established, you relax both the aeleron and
rudder and PUSH forward on the stick same as you would normally pull
back in a turn, (just enough so that the "+" remains the same height
below (above) the horizon) and the nose will slide along parallel to
the horizon (downwards to the right) towards B, where you then apply
RIGHT aeleron and LEFT rudder and ease off the forward pressure as the
wings come back to level. It's a lot of fun! You can try this out
(without the rudder) on the flight simulator, and you will see that
you get the hang of it pretty soon. Try to do a low inverted pass
over the runway, making corrective turns to stay right over the
centerline! THAT is not so easy!
--
(O)((O))(((O)))((((O))))(((((O)))))(((((O)))))((((O))))(((O)))((O))(O)
(O)((O))((( sle...@park.bu.edu )))((O))(O)
(O)((O))((( Steve Lehar Boston University Boston MA )))((O))(O)
(O)((O))((( (617) 424-7035 (H) (617) 353-6741 (W) )))((O))(O)
(O)((O))(((O)))((((O))))(((((O)))))(((((O)))))((((O))))(((O)))((O))(O)