As with my other AP topics here again I only use two AP features:
- HDG Hold
- VS Hold
… and adjust the airspeed manually … and I do not set any target altitude.
I have summarize this as "AS VS Hold get ignored (is impacted) by airspeed changes" because it seems to come in many different flavors … but it feels to be an unintended interaction between the airspeed and other AP aspects.
In this video … https://www.objectfarm.org/Public/Misc/2022.11.22-MSFS-MV22B-Bug-AP-VS_descent_on_airspeed_change.mp4
- I set a rapid -1300 ft VS descent
- and reduced the throttle to 79% NR … to not overspeed during the descent
- ... as a consequence the airspeed slowly drops from 250 to 178
- but during the process the AP slows raises the VS rate
- ... which in the end reaches around -2000 ft
The MSFS forum has reports about other incidents which match some of my other (similar) observations … e.g. that some throttle increases will result in extremely high climb rates (which override the configured VS Hold)
The strange thing is that in the past I had no problems to have stable (VS = 0ft) AP flights with airspeeds as low as 140 ktas. So it is not that 180 ktas is a hard lower limit below which the AP goes into a dive to regain or hold the airspeed.
Miltech Simulations @MiltechSimulationsThis may have to do with the other thread regarding V/S PID Control. I’m prioritizing the other fix, and then I’ll look into how this may get impacted.
Autopilot is being a headache. The behaviour is not as expected, though its using the stock MSFS AP. Modeling the system from scratch may be the “best solution”, though by far the hardest. Autopilot systems work as a closed-loop feedback control system (which reminds me of painful times during my Mech Engg major..)
I’m currently doing testing with PID values from other stock aircraft and seeing how the Osprey behaves.
Miltech Simulations @MiltechSimulationsTested the new autopilot PID under these conditions - it does handle better, but at speeds under 150kts the aircraft does increase V/S. With the new AP, I'm getting -2000fpm at 135kts, which sounds more reasonable, given that this is getting close to the stall speed. It would be expected to start conversion at around 170kts.
I would assume that the AP is increasing V/S to keep the aircraft well above stall speeds.
@nenenuiI will take a look at the new version … and discard my additional AP feedback for now, as it might (hopefully) be obsolete.
- Miltech Simulations @MiltechSimulations
The files have been submitted to ORBX, update should be out in a few hours.