OK… back to some pretty Basic stuff.
There is Real World, and there is the SIM’s World which is trying to Simulate the Real World, the best it can, given certain limitations. Over the year of FS development, many of those limitations have been overcome.
I am working on a “Better” KAP140 AP for the C172 and other small GA planes, and have hit some obvious issue, as far as a correct simulation is concerned.
Lets not jump the gun, ( this issue is most likely Historically the result of Spring Centering Joysticks & Yokes )— so before getting into how the AP is operating INCORRECTLY in MSFS, it would be better to get a clearer understanding how it operates in the Real World.
Let just deal first with Vertical Control.
There are two basic ways this is done:-
(1) Control the vertical Trim, with a Motorized trim wheel (Airlines and large GA planes)
(2) Control the Yoke, with a servo, applying pressure to the yoke, (and NOT altering the trim with the AP) (Small GA planes)
The C172 should be using method #2 … (It should NOT have a motorized Trim Wheel)
So, The AP should NOT be altering the trim … only the Pilot can do that, by physically moving the trim wheel.
When the AP is turned on, it decides how much pressure to apply to the Yoke, to obtain the desired VS, and start to push or pull on the yoke.
If it required too much force, it disengages.
If the force required is within its limits, it applies that force, and if that force is not minimal, it indicated on the AP, that the plane is out of trim, and it is up to the Pilot to mechanically trim the aircraft, so that the AP’s force is reduced, and it stop indicating an “out of trim” condition.
So why, in MSFS, is the C172 AP modeled like an Airline AP, and not a small GA AP.
Because it has always been done that way … probably initially, because of the issue of the Joystick/Yoke, being spring loaded to fixed center position.
One obvious result of this, is if the AP goes “CRAZY” it can unrealistically spin the Trim wheel to its limits !!!
So – now my 2 question are:
(1) Is the above analysis correct ?
(2) IF so, is there a better way to Simulate a GA AP, given the limitations of Fixed center Controllers.
Hint: Just because the controller is sprung to center, does NOT mean that the AP software has to take that input directly — it could be OFFSET in the Software to “simulate” that the Centered Joystick was not actually at it’s spring loaded center. (and that offsetting software could be inhibited for Force Feedback Joysticks)