elevator responses

[kirovswm]

I posted this question in the a2a forums as well, as I'm not sure if this addon or their modeling software should be making the correction.

https://a2asimulations.com/forum/viewto ... 07&t=53662

I have a question about how the a2a c172 responds to elevator changes with the IRIS dynamic yoke. I am using Prepar3d v3.2 and FSForce addon. For this example, assume the following. FSUIPC has a value of -16,000 for full back deflection, +16,000 for full forward defelction, and 0 for neutral.

Most spring yokes when you pull or push on the yoke, the springs return it back to neutral, so the elevator value rests at 0 in cruise.

However, the IRIS dynamic yoke has the ability to stay where it is after trimming through magnetic induction. So for example, if in level flight I have the yoke pushed forward and I have a value of +3,000 for the elevator, I'm still able to trim the plane to keep it level, even though the game has the yoke registered as being pushed forward a little, and the pressure in the yoke goes away, like in real life.

My observations have been, that when I want to change altitudes, or apply a small amount of back pressure when I bank, I feel the elevator response is much more drastic than in a real c172 when I am trimmed and my yoke is not at neutral. The VSI goes very quickly to around 1,000feet/min, and the artificial gyro quickly goes to 10 or 15 degrees with only a little bit of pressure.

My question is, is this software modeled to always treat the elevator response as a delta from 0? A change from +3,000, but trimmed, to +4,000, the game "feels" like it wants to make an elevator change of+4000 instead of +1000, according to the behavior described in the above paragraph. It would make sense since most yokes are spring based and naturally return to 0. This is a problem though for the IRIS yoke if this is true.

Is my guess correct?

[RussDirks]

I think what you are describing is an over sensitivity in the pitch axis to control inputs, correct? I think there are two things going on here:

1. Aircraft flight dynamics. I find that in general, aircraft in FSX/P3D are too sensitive in the pitch axis. One way to deal with that is to hand edit the "elevator_effectiveness" and "pitch_stability" values in the aircraft.cfg file. You can find them under the [flight_tuning] section.
2. Null zone. In FSX/P3D, the null zone is assumed to be in the center of travel, but with FS Force, it is possible to have the aircraft trimmed with the yoke in any position. In the example you mentioned above, you had the aircraft trimmed with the yoke pushed forward. What that means is that you are no longer in the "null zone" as defined by FSX/P3D, so even the slightest change you make in the controller position will result in elevator movement, which will affect the pitch attitude of the aircraft.

[kirovswm]

So in scenario 2, let's say I'm trimmed in level flight, and fsuipc says my yoke is at +2000. There is no null zone here, which makes sense. But if I push forward slightly, and fsuipc now says my yoke is at +3000, which has caused me to leave level trimmed flight, my question is, is that modeled as if I'm asking the simulator to behave as if I've gone from 0 to +3000? An elevator change of a value of 3000 will make the airplane pitch more drastically than an evelevator change value of 1000.

That's how the simulator feels to me at least. Kind of like when in autopilot, the hardware value for the trim wheel does not change because it hasn't been physically moved, but the trim in the game has been changed by the autopilot. When you leave autopilot, and try to move the trim wheel physically, I've had the yoke shoot back at me and the airplane pitch straight up because of the difference in physical controller value, and software trim value.

Does that make sense how I've worded it?

[RussDirks]

So in scenario 2, let's say I'm trimmed in level flight, and fsuipc says my yoke is at +2000. There is no null zone here, which makes sense. But if I push forward slightly, and fsuipc now says my yoke is at +3000, which has caused me to leave level trimmed flight, my question is, is that modeled as if I'm asking the simulator to behave as if I've gone from 0 to +3000?

No, not by FS Force.

I would recommend not programming your elevator axis through FSUIPC ... just set it up through P3D. See if that makes any difference. If you are using the AP Follow feature of FS Force, you cannot use FSUIPC for the pitch axis, as it conflicts with FS Force, and if you are using the Aileron Remapping feature of FS Force, you cannot sue FSUIPC for the aileron axis, for the same reason.

Are you getting this same sensitivity with other aircraft as well?

[MarkH]

Russ, I am wondering if this is due to FSX managing the pitch axis in a non-linear way. In other words, exactly like the problem you have tackled with aileron re-mapping, but on the pitch axis. I understood that FSX did this for all analog axes. I tried checking this out by loading up the default C172. It's hard to tell from the VC yoke movements but in the external (spot) view I would say the elevators are moving in a similar exponential way with yoke deflection.

[RussDirks]

MarkH,

Yes, FSX seems to be doing some type of "exponential mapping". Control yoke movement near the center of travel seems to cause less movement of the elevator (observed from spot view) than similar movement of the yoke when it is towards the extreme edge of travel. With my A2A 182, however, the elevator motion seems linear across the full range of control yoke movement.

[MarkH]

My observations have been, that when I want to change altitudes, or apply a small amount of back pressure when I bank, I feel the elevator response is much more drastic than in a real c172 when I am trimmed and my yoke is not at neutral

Not sure if the obvious question has been asked - if you are using FSUIPC, are you using a 'slope' on the elevator axis? You can't do that, for the same reason you can't use a null zone. But if you do, I would have thought it has the potential to give the kind over-sensitivity you are describing, at least in certain trim positions.

[lfthomaz]

Hello Russel,
I'm having the same problem as reported by Mark, way too much sensitivity on the elevator (with a MS Force Feedback 2 and A2A C172).
If I set a high value for the null zone, it's just ok when the joystick is centered (with high values of trim, it gets really difficult to control).
Can you share your settings so I can try to get a more stable experience (Prepar3d control setings, aircraft.cfg for A2A airplanes, etc)? Do you recommend using FSUIPC?
I really like your software (and just bought it) but I want to get the max out of it!
Thanks a lot!

[RussDirks]

No, there is no need to use FSUIPC. You're best bet is to decrease elevator_effectiveness and increase pitch_stability in the aircraft.cfg file until you get it the way you like it.

[bucastef]

My elevator is very sensitive as well. I changed the cgf file and decreased the elevator effectiveness to 0.5 and increased the pitch stability to 1.5, but it does not seem to correct the problem.
One other thing I noticed is that while the yoke movements seem to linearly correlate with the virtual yoke, while looking at the control surfaces of the airplane I get almost full deflection of both ailerons and elevators within 1/3 of yoke travel. Is this normal?
I am running FSX with iris dynamics yoke, and I have the aileron remapping enabled.
Any suggestions?
Thanks
Stefan

[RussDirks]

One other thing I noticed is that while the yoke movements seem to linearly correlate with the virtual yoke, while looking at the control surfaces of the airplane I get almost full deflection of both ailerons and elevators within 1/3 of yoke travel. Is this normal?

No, that seems a bit extreme. Is the yoke properly calibrated? It's been awhile since I've used my Iris yoke so I can't quite remember if it has it's own calibration routines, or if you do it though Windows Game Controllers.

Don't be afraid to be more aggressive with the settings in the aircraft.cfg file. Keep increasing/decreasing the values until you can actually detect a difference.