fly/fgdata
1
0
Fork 0
Code Activity

Maik JUSTUS: YASim helicopter FDM update

Browse source
This commit is contained in:
mfranz 2006-08-14 22:38:59 +00:00
parent cd74bc8572
commit eeeea44805
2 changed files with 145 additions and 23 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

166
Docs/README.yasim
View file

@ -441,9 +441,12 @@ control: This element, which can appear in two different contexts,
rotor: A rotor. Used for simulating helicopters. You can have one, two
or even more.
There is a drawing of a rotor in the Doc-directory
(README.yasim.rotor.gif) Please find the measures from this drawing
for several parameters in square brackets [].
If you specify a rotor, you do not need to specify a wing or hstab,
the settings for approach and cruise will be ignored then. Instead
stored results from the c182 will be used.
the settings for approach and cruise will be ignored then. You have
to specify the solver results manually. See below.
name: The name of the rotor.
(some data is stored at /rotors/name/)
@ -453,21 +456,44 @@ rotor: A rotor. Used for simulating helicopters. You can have one, two
All angles are in degree, positive values always mean "up".
This is not completely tested, but seem to work at least
for rotors rotating counterclockwise.
A value stall gives the fraction of the rotor in stall
(weighted by the fraction the have on lift and drag
without stall). Use this for modifying the rotor-sound.
The torque property has a bug.
x,y,z: The position of the rotor center
nx,ny,nz: The normal of the rotor (pointing upwards, will be
normalized by the computer)
fx,fy,fz: A Vector pointing forward, if not perpendicular to the
normal it will be corrected by the computer
diameter: The diameter in meter
diameter: The diameter in meter [D]
numblades: The number of blades
weightperblade: The weight per blade in pounds
relbladecenter: The relative center of gravity of the blade. Maybe
not 100% correct interpreted; use 0.5 for the start and
change in small steps
change in small steps [b/R]
chord: The chord of the blade its base, along the X axis
(not normal to the leading edge, as it is
sometimes defined). [c]
twist: The difference between the incidence angle at the
blade root and the incidence angle at the wing
tip. Typically, this is a negative number so
that the rotor tips have a lower angle of attack.
taper: The taper fraction, expressed as the tip chord
divided by the root chord. A taper of one is a
bar blade, and zero would be a blade ending
at a point. Defaults to one. [d/c]
rel_len_where_incidence_is_measured: If the blade is twisted,
you need a point where to measure the incidence angle.
Zero means at the base, 1 means at the tip. Typically
it should be something near 0.7
rel_len_blade_start: Typically the blade is not mounted in the
center of the rotor [a/R]
rpm: rounds per minute.
ccw: determines if the rotor rotates clockwise (="false") or
counterclockwise (="true"), (if you look on the top of the
normal, so the bo105 has counterclockwise rotor)
ccw: determines if the rotor rotates clockwise (="0") or
counterclockwise (="1"), (if you look on the top of the
normal, so the bo105 has counterclockwise rotor).
"true" and "false" are not any longer supported to
increase my lifespan. ;-)
maxcollective: The maximum of the collective incidence in degree
mincollective: The minimum of the collective incidence in degree
maxcyclicele: The maximum of the cyclic incidence in degree for
@ -478,19 +504,76 @@ rotor: A rotor. Used for simulating helicopters. You can have one, two
the aileron like function
mincyclicail: The minimum of the cyclic incidence in degree for
the aileron like function
airfoil_incidence_no_lift: non symmetric airfoils produces lift
with no incidence. This is is the incidence, where the
airfoil is producing no lift. Zero for symmetrical airfoils
(default)
incidence_stall_zero_speed:
incidence_stall_half_sonic_speed: the stall incidence is a function
of the speed. I found some measured data, where this is
linear over a wide range of speed. Of course the linear
region ends at higher speeds than zero, but just
extrapolate the linear behavior to zero.
lift_factor_stall: In stall airfoils produce less lift. Without
stall the c_lift of the profile is assumed to be
sin(incidence-airfoil_incidence_no_lift)*liftcoef;
And in stall:
sin(2*(incidence-airfoil_incidence_no_lift))*liftcoef*...
...lift_factor_stall;
Therefore this factor is not the quotient between lift
with and without stall. Use 0.28 if you have no idea.
drag_factor_stall: The drag of an airfoil in stall is larger than
without stall.
Without stall c_drag is assumed to be
abs(sin(incidence-airfoil_incidence_no_lift))*dragcoef1...
..+dragcoef0);
With stall this is multiplied by drag_factor
stall_change_over: For incidence<incidence_stall there is no stall.
For incidence>(incidence_stall+stall_change_over) there is
stall. In the range between this incidences it is
interpolated linear.
The airfoil of the rotor can be described in two ways. First you
can define the needed power for different pitch values and the
total lift force at a user-defined pitch value. Don't use pitch
values greater than the stall incidence. You could get strange
results.
pitch_a: A collective incidence angle, used for the next token
forceatpitch_a: The force, the rotor is producing when the incident
angle is equal pitch_a. I.e. hover-pitch and a force
angle is equal pitch_a. Without ground effect and with
maximum translational lift. I.e. hover-pitch and a force
equivalent to the weight. (in pounds of force)
pitch_b: A collective incidence angle, used for the next token
poweratpitch_b: the power the rotor needs at pitch_b. (i.e. at the
bo105 the main rotor consumes bout 90% of the engine power,
and 9% the tail rotor. In kW. Used for calculation of the
torque.
and 9% the tail rotor. In kW.
poweratpitch_0: the power the rotor needs at zero pitch.
In kW. Used for calculation of the torque.
notorque: If set to "true" the calculated torque is always zero.
Very helpful while adjusting rotor parameters.
In kW. Used for calculation of the airfoil coefficients.
In near future you can define them directly.
The second way is to define the lift and drag coefficients directly.
Without stall the c_lift of the profile is assumed to be
sin(incidence-airfoil_incidence_no_lift)*liftcoef;
And in stall:
sin(2*(incidence-airfoil_incidence_no_lift))*liftcoef*...
...lift_factor_stall;
Without stall c_drag is assumed to be
abs(sin(incidence-airfoil_incidence_no_lift))*dragcoef1...
..+dragcoef0);
See above, how the coefficients are defined with stall.
The parameters:
airfoil_lift_coefficient: liftcoef
airfoil_drag_coefficient0: dragcoef0
airfoil_drag_coefficient1: dragcoef1
I read in a forum, that, if you calculate the lift of an heli rotor,
you will get a value larger than the measured one. This seems to be
valid for this simulation. If you use values for the lift
coefficient from real airfoils you will get unrealistic high lift as
result (approx. a factor of 2). As starting parameters you can use
airfoil_lift_coefficient="1.9"
airfoil_drag_coefficient0="0.0075"
airfoil_drag_coefficient1="0.2"
flapmin: Minimum flapping angle. (Should normally never reached)
flapmax: Maximum flapping angle. (Should normally never reached)
flap0: Flapping angle at no rotation, i.e. -5
@ -506,7 +589,7 @@ rotor: A rotor. Used for simulating helicopters. You can have one, two
rotors) For rotors without flapping hinge (where the blade
are twisted instead, i.e. Bo 105, Lynx) use a mean value,
maybe 0.2. This value has a extreme result in the behavior
of the rotor
of the rotor [F/r]
delta3: Some rotors have a delta3 effect, which results in a
decreasing of the incidence when the rotor is flapping.
A value of 0 (as most helicopters have) means no change in
@ -522,17 +605,56 @@ rotor: A rotor. Used for simulating helicopters. You can have one, two
If you know the flapping angle for a given cyclic input you
can adjust this by changing this value. Or if you now the
maximum roll rate or ...
translift: Helicopters have "translational lift", which is due to
turbulence and hard to calculate, so this simulation uses
a phenomenological ansatz. Use .1 for the start value
dragfactor: The drag of the rotating rotor perpendicular to the
rotor plane is larger than the drag of the not rotating
rotor. Hard to calculate, so it is added phenomenological
translift_maxfactor: Helicopters have "translational lift", which
is due to turbulence. In forward flying the rotor gets less
turbulence air and produces more lift. The factor is the
quotient between lift at high airspeeds to the lift at
hover (with same pitch).
translift_ve: the speed, where the translational lift reaches 1/e of
the maximum value. In m/s.
ground_effect_constant: Near to the ground the rotor produces more
torque than in higher altitudes. The ground effect is
calculated as
factor = 1+diameter/altitude*_ground_effect_constant
number_of_segments: The rotor is simulated in four different
directions (probably this will be extended in future).
In every direction the rotor is simulated at
number_of_segments points. If the value is to small, the
rotor will react unrealistic. If it is to high, cpu-power
will be wasted. I now use a value of 10, but probably a
smaller value for the tail-rotor would be sufficient.
Any rotor needs a <control> subelement for the engine
(ROTORENGINEON) and can have <control> subelements for the cyclic
All rotor can have <control> subelements for the cyclic
(CYCLICELE, CYCLICAIL) and collective (COLLECTIVE) input.
rotorgear: If you are using one ore more rotors you have to define a
rotorgear. It connects all the rotors and adds a simple engine.
In future it will be possible, to add a YASim-engine.
max_power_engine: the maximum power of the engine, in kW.
engine_prop_factor: the engine is working as a pd-regulator. This
is the width of the regulation-band, or, in other words,
the inverse of the proportional-factor of the regulator.
If you set it to 0.02, than up to 98% of the rotor-rpm
the engine will produce maximum torque. At 100% of
the engine will produce no torque. It is planned to use
YASim-engines instead of this simple engine.
engine_accell_limit: The d-factor of the engine is defined as the
maximum acceleration rate of the engine in %/s,
default is 5%/s.
max_power_rotor_brake: the maximum power of the rotor brake, in kW
at normal rpm (most? real rotor breaks would be overheated
if used at normal rpm, but this is not simulated now)
yasimdragfactor:
yasimliftfactor: the solver is not working with rotor-aircrafts.
Therefore you have to specify the results yourself.
10 for drag and 140 for lift seem to be good starting
values.
The rotorgear needs a <control> subelement for the engine
(ROTORGEARENGINEON) and can have a <control> subelement for the
rotor brake (ROTORBRAKE).
The rotor simulation is very "beta" and not finished yet. So don't
spend too much time to adjust a flight behavior to the smallest
details now.

BIN
Docs/README.yasim.rotor.png Normal file
View file

Binary file not shown.
Side by side

After

Width:  |  Height:  |  Size: 7.9 KiB