Remember that adverse yaw is momentary, with the exception of when you have to hold aileron throughout the turn due to overbank tendency or the opposite, lateral stability that wants to return you to wings level.
After the roll-in stops, given ailerons neutral, any remaining slip or skid is just uncoordinated flight.
The indication of adverse yaw should be momentary ball in the direction of the turn (slip), possibly a slight indication of turn in the opposite direction, very briefly, if it’s a turn and slip indicator. The gyro in a true turn coordinator is canted 30° so it picks up the gyroscopic precession induced by the roll, as well as the yaw, as well as the turn, so the roll should override any opposite turn indication.
So in real life, as soon as you are using the ailerons at roll-in you are using the rudder simultaneously, when ailerons going level you are releasing the rudder, unless slip or skid appears.
Sort of - yes, you coordinate the turn with rudder. How much rudder depends on which way you’re turning, your indicated airspeed, the rate of roll, and of course the design of the aircraft itself.
A single-engine prop aircraft often provides enough left yaw that you don’t need a ton of left rudder entering a left turn. You will often need more right rudder into a right turn. Slow down and the effect is magnified for each direction.
As far as releasing it, you just use enough that the ball stays centered. Remember left turning tendencies, wind gusts, maintaining the bank angle, speed, and altitude are all things that can affect the quality of the turn, so it’s a constant game. In a perfect world without all those little imperfections, though, it should be fairly close to center with little rudder pressure once you’ve established the bank angle and do not have any aileron still in.
I was flying the Curtiss Jenny yesterday; its instrumentation is bare-bones so there’s no turn/slip indicator, but I could definitely see by eye the effect of adverse yaw when using vs not using a little rudder.
During the roll, without rudder I see that the aircraft yaws a bit in the opposite direction of the roll, then afterwards starts turning as expected – while its heading remains more stable during the roll if I coordinate with the rudder.
Nicely done to the devs on that one.
(Also, I highly recommend a head tracker like TrackIR for those open-cockpit biplanes!)
The Curtis Jenny actually is one of the aircraft with a very high fuselage_later_cx values you can see here.
This is a drag coefficient value for the fuselage and Asobo recommondation was 0.2 - 0.8 and a default of 0.4, which is very low and of the possibilities why we don’t see much adverse yaw and uncoordinated turns in general.
The community suggested 0.8 - 1.3.
The CJ4 has this value set to 3, which is ridiculously high even for that airplane and it comes with other unrealitic side effects like enormous slip
I’m thinking about expanding this thread to something like ‘‘adverse yaw / slip / skid’’
and integrating the fuselage_cx value, as this is very much overlooked by developers and wrong values are suggested by Asobo in their SDK.
At the end this list should simply provide an information on which aircraft you can use the rudder pedals realistically.
updated the list with the fuselage_lateral_cx values with help of the bug report section.
As you can see most developers are following the recommendation of 0.4 - 0.8,
while most of them using the default of 0.4.
0.4 equals the drag of a sphere and is much too low for most airplanes.
While 0.8 equals the drag of a cylinder and should be used as a default starting point for developers.
Maybe that’s one of the reasons we can easily do ‘‘uncoordinated’’ turns in MSFS while beeing completely coordinated.
Next thing I want to do is further looking into the wing aspect ratio and how wing roll rate is simulated in MSFS.
But it’s most likely a airplane geometry calculation result of the ‘‘modern flightmodel’’.
Again, it depends on the plane and roll rate and direction. A slow roll to the left in a 172 may not need much rudder at all. I haven’t tested enough of this in the sim to determine what does and doesn’t, specifically.
I’m always using rudder and working to keep the ball centered. Are you talking about in the roll or in the turn?
One thing to also remember is that real-world rudders in GA planes don’t exactly self-center. They also have trim tabs (fixed for 172, adjustable for Cherokees) and torque, slip, etc, increases pressure and pushes the rudder to whatever AoA it’s trimmed, which isn’t necessarily a dead-center feeling. It will often fight against your foot pressure, which is a cause of uncoordination if left unattended. So our self-centering rudder pedals do us a little injustice, similar to some of the issues we run into with yokes, etc.
That said, I put Velcro strips inside my rudder pedals, under the rollers, so they don’t automatically center. I can still find/feel the center, but there’s quite a bit of pressure and they can stick, making me work harder.
BTW, this is why I always say that we shouldn’t be getting overly technical when comparing real-world to sim. There are a bunch of factors that the sim is completely unable to replicate without kinda faking it, and it causes a lot of misconception.
That’s weird, I never had to use even a little bit of rudder since I’m paying attention to the turn indicator. The ball maybe moves very very shortly out of center when using the ailerons for the roll, but it selfcenters instantly and stays there for the rest of the turn, so it’s better to not use any rudder at all.
Wow, I didn’t know that, I thought only in helicopters the rudders doesn’t self center. Thanks for the info!
It’s usually returns close to center, but there’s perceptible pressure depending on flight condition. The aerodynamics do most of the work re-“centering” it, like the ailerons and elevator, but a little deflection goes a long way and trim plus changing conditions will mess with it a bit.
The last major hurdle in sim fidelity is simulated, realistic aerodynamic feedback. We can’t even really get past some workarounds until that happens. And even then, you still have to contend with all the folks who can’t/won’t have feedback.
Consider all the people who think the airplanes are too squirrelly, while they subject them to massive swings in pitch because they’re using a gamepad and pulling the elevator from stop to stop in roughly an instant. That’s pretty much not possible irl due to aerodynamic feedback and control throw distance in real planes.
Or the anti-servo tab in real-world Cherokees adding feedback at the extremes of travel, causing less propensity to over-control. Can’t simulate that at all, the best we get is a workaround.
Ok maybe I’m also having a misconception of the need for rudder for turns.
I believed small GA’s always have some kind of inbalance while making a turn and you have to compensate that with small rudder input. But outside of adverse yaw it’s probably not often the case.
coming from simracing into ‘‘simflying’’ the missing force feedback is still a big bummer for me
technically force feedback should be possible, but as far as I know it’s a licensing problem.
It’s that, but even if we had it, a relatively small percentage of users will want to/be able to afford it, and mostly only if it’s scaled enough to get it down to a reasonable price range. There’s going to be a lot of chicken/egg if/when that ever happens. Plus everybody will probably have to go back and re-write all their code, haha.
They can and often do. It’s just tied into so many other factors that the answer is always going to be “it depends.”
I’m not exactly sure how the total drag of the wings is figured, but I know for sure that the aspect ratio of the wing and oswald efficiency factor are inversely proportional to the induced drag. Coefficient of lift (based on AoA in the lift_coef_aoa_table) is also part of the equation for induced drag (more lift = more drag). Pretty sure the sim follows this by the book.
I’m assuming parasitic drag is covered at least partially by drag_coef_zero_lift and geometry figures but that applies to the aircraft as a whole.
drag_coef_zero_lift
…In the modern flight model, this defines the target Cdo that will be distributed over all the surfaces of the aircraft when building the airplane used in the aerodynamic surface simulation. Once the aircraft is built, it will then be normalized to match exactly the target Cdo.
So this number should also have an effect on adverse yaw and slip/skid behavior.
But we can’t check anything here, we have to trust that it works as intended and developers using the right value in combination with correct airplane geometry. But at least it indicates that it’s simulated to begin with .
On a side note:
the possible wrong Fuselage_lateral_cx value suggestions got forwarded to the MSFS flight model people and will be reviewed.
i was able to cobble together FFB pedals for $300 (not including the pedals themself). Not cheap but not brunner $$$$ either. the cable-pull design could be adapted to most pedals.
If I’m not wrong there is no separate drag coef for the floats,
so wouldn’t it make sense if developer are using the fuselage_lateral_cx value to simulate that?
It’s simply just putting in a slightly higher number just like WB-Sim did with the 172 float version.
Meanwhile the Asobo 172 using the same number for every version, so you have basically no difference in flight behavior.
