#29 Twin (C-310) Aerodynamics; Engine-Out
It very well could be a model setup issue but with the C-
310 after taking off, climbing to a comfortable altitude
and speed, and chopping the throttle on the right
engine, I slowly pitched up to bleed off speed little by
little. As the speed bled off, I held my heading with
rudder and kept the wings level with aileron.
From the readme:
Minimum single-engine control speed (Vmca): 75 KIAS
However, I was able to fly right through this until I got
the stall horn, (about 60 kts?) and all the time, the
rudder had plenty of effectiveness to hold heading. In
the real thing (assuming the README is correct) at
about 75 knots the rudder loses enough effectiveness to
hold heading against the one good engine at full throttle
and you begin an uncontrollable yaw. This doesn't
happen right now in the JSBSim C310 anyway.
I'm sure this is just a matter of tweaking the
configuration file. But this is an important behavior to
have reasonably correct in small twins.
The thrust from the good engine is only half the
asymmetry -- the other half is the drag from the
windmilling engine (until the pilot feathers the propeller).
A NACA report for a light twin suggests 0.103 for
Cn_beta.
Cn_beta for some aircraft (per rad):
Navion: 0.071 (Raymer ?)
C-172p (JSBSim, from Raymer):
-0.349 -0.0205
0 0
0.349 0.0205
This is roughly 0.06.
Cherokee (McCormick): 0.067
C-310 (JSBSim): 0.1444
This is twice as high as the other aircraft. It could be
due in some measure to a larger vertical tail, but I
wonder if perhaps this value is too high? When coupled
with the correction of drag due to prop, then I suspect
we'll be a lot closer.
I tried to fix this problem in JSBSim a year or two ago,
and I seem to recall that no one on the flight model list
could quite figure out how to code it back then. I also
took a stab at Yasim, and failed just as miserably.
Neither model is set up to have the propeller driving the
engine rather than the engine driving the prop.
The rule of thumb for pilots is that a windmilling propeller
creates as much drag as a disc of the same size, but
that's too vague for modelling (plus, it doesn't handle
the partial-windmilling situation). What we need to
figure out is how much drag we get from the propeller
turning the crankshaft, compressing the cylinders, and
spinning the accessory drives (vacuum pump, alternator,
etc.).
Logged In: NO
Vmca decreases with increasing altitude so that above a
given altitude you will get stall warning before you see the
nose slice. This is especially true in Albuquerque where
density altitudes during cruise can easily exceed 10,000 feet.
It's not uncommon for multi-engine instructors to demonstrate
a pseudo-Vmca by using their feet to limit the amount of
rudder the student can apply just so they can see what the
nose slice looks like.
Also, the parameter you want to look at is not Cn_beta, aka,
weathercock stability. For this, you want to look at
Cn_delta_rudder.
dgsharpes@gbronline.com