Major autopilot issues

  • Sometimes the autopilot will start acting up and just go full left or right aileron when engaged. This can happen regardless of heading mode being on or off, and regardless of which LNAV mode is selected (e.g. hdg, nav, loc, etc). It usually starts working again after a minute or so, but no amount of fiddling (switching AP switches on/off, adjusting nav mode, etc) would fix it.

    Perhaps a related issue is that when AP is on but heading mode is off, turning the nav selector to a different mode causes the yoke to twitch. But if heading mode is off that knob should do absolutely nothing (if my understanding is correct).

    I also had an issue where I would engage the AP but not heading mode, and the plane would start turning ever so slowly rather than holding the wings level. This is more than just drift or wind, the attitude indicator was showing about a constant 2 degree bank. At first I thought I'd maybe adjusted the roll knob by mistake, but even after turning it significantly to the right, the plane kept banking about 2 degrees left.

    On one occasion when the plane was going hard right for no apparent reason (but not hard enough to start diving), I waited to see if it was searching for something. But it never stopped turning and I watched it do more than two full revolutions.

  • JF Staff

    Thanks for the comprehensive feedback! We are currently tweaking the autopilot logic, although that can be challenging given the quirks of the MSFS autopilot system, and hope to release a new update shortly.

    You're correct that the navigation coupler knob should be inoperative if the heading switch is set to OFF.

  • The AP is acting up when you apply some aileron input while its active.
    I recorded a small video:

  • Hi, I recognize that randomly by engaging the AP (also with HDG on/off switch in off) the ailerons goes in opposite direction and if happends during flight the plane goes in a spin.

    This tested also at ground and plane powered off:
    If I switch AP switch only the ailerens goes in opposite directions.

    Piper just updated at 0.3.0

    Thanks for suggestion

  • While I have not thoroughly researched the operation of the actual autopilot used in this aircraft, when turned ON without HDG or NAV mode engaged, most autopilots will default to either wings level mode, or simply hold whatever bank you had when you turned it on... This may be the default autopilot behavior in MSFS.

    The correct way to engage the autopilot in most aircraft is to first set your heading bug to the current heading, then engage HDG mode, then engage the autopilot by turning it ON or engaging CMD mode.

  • @RetiredMan93231
    If this happends I disengage the AP and proceed the flight manually without problems. But cannot use the autopilot.
    At ground with engine and battery off if I switch to AP Engage switch ON the ailerons goes opposite full direction...if I Disengage AP switch the ailerons goes to neutral normal position. Does not matter if the HDG switch is off or on and the position of the AP mode knob.

    The only way to reset this is to restart the flight.

  • This post is deleted!

  • To test the autopilot operation while on the ground...

    • Set the heading bug to the current heading.

    • Engage HDG mode.

    • Engage the autopilot ON

    • Change the heading bug to the left and right of the current heading and verify that the yoke turns in that direction.

  • I may have figured out whats going on, at least partially. (v0.8.0)

    When the autopilot is acting up, I've been able to repeat the following series of events a few times now:

    • AP on, HDG off, plane banks hard right for seemingly no reason. It will continue to fly in circles as long as you pull back on the yoke to maintain altitude. Doesn't stop at the heading bug or cdi. =
    • WIth HDG switch off, the roll dial doesn't have any effect. Plane just keeps doing whatever it wants
    • Toggle HDG switch, move cdi/bug around, behavior continues
    • Turn on HDG mode and leave it on until it captures the heading bug.
    • At this point, the weird behavior ends. I can now adjust the heading bug and it will turn gradually to follow the selected heading. Turning hdg mode off makes it go wings level, and the roll dial is working again too.

    So my theory is this: when the AP gets into this state, it thinks it's looking for the heading bug (or cdi depending on the selected mode) but if the HDG switch is off, it can't actually capture it because that functionality is only supposed to work if the HDG switch is on. So it keeps turning in circles.

    It seems there is some blocking code whose prime directive is "find the heading bug", but when it does find it, the code that is responsible for leveling off says "no can do, HDG switch is off".

    Similar behavior can be observed with the HDG switch off and NAV mode selected on the coupler knob. Tune a VOR, and even with the HDG switch off, it will fly the course selected by the cdi. Adjust the cdi a bit, and it will turn to follow it.

  • Some more observations:

    With AP switch on, HDG switch off, tuned to a VOR and starting on course.

    The yoke stays level as expected, but the ailerons are moving as they would in NAV mode. Turning the roll knob seems to make these two mode "fight" each other. After turning the roll knob past a point, it finally wins out and the plane banks, but after returning the knob to neutral, the plane banks back hard, presumably in an attempt to re-capture the course.

    I'm currently working on recording a video that showcases most of these issues happening but isn't too long, I'll post that here once I've got it

  • Was going to make a more organized video but every time I tried to do something, another new thing went wrong so here is a 5 minute video showing the main issue and some timestamps to jump to listed below.

    0:00 At the start, I have set up a course on the GPS, and I am following it. No wind and balanced weight to make things easier. importantly, the HDG switch is OFF.

    0:08 As I adjust the CDI to the left or right slightly, you can see the plane turn in that direction. I then let it re-align with the course.

    1:55 Then I adjust the roll command knob several clicks to the left and it works as expected.

    2:21 Now, messing with the CDI doesn't cause the plane to try and follow the new course. Good.

    2:28 Then I turn the roll command knob two or three clicks back toward center (but still very much left of center) and the plane banks HARD right. Despite, as you can see in the video, the yoke still being turned to the left, as expected. At this point, it's clear the plane is following the NAV course while the yoke is correctly obeying the roll command.

    2:38 Shortly after this I turn the knob further left and the plane banks left again and seems to be obeying the roll command as well, though I'm not sure if that is a direct result of me turning the knob again.

    3:35 Then I try to re-center the roll knob, and the hard bank happens again. The plane turns and slowly levels out at seemingly a random heading (not the HDG bug or the CDI, not even a direction which would indicate an attempt to capture the course).

    4:28 Another attempt to get the roll command to work, resulting in a hard bank yet again. If you notice here, the yoke is almost full left, but the ailerons are neutral.