Autopilot smoothness and questions
-
Hey Nick, first of all, I’ve only had a couple of hours today to fly but well done on releasing such a fully featured and quality product….. a rare thing these days from many developers. It’s going to bring many hours of enjoyment.
Q1. One of the only negatives so far is the relative jerky and abrupt behaviour of the autopilot. Altitude captures and even small changes of commanded VS produce an unrealistic abrupt pitching movement. Is there any to smooth this out while still maintaining responsiveness and accuracy?
Q2. Since PITCH mode is actually the only smooth acting mode, is there any way to select this mode directly from ALT HOLD?
Q3. (Unrelated pedantic question…..IRL manuals show a LIN (linear) flag in top left ND for FMS nav source. Was this changed in later avionics? It’s not present).
Thanks.
-
Thank you so much. I'm so glad you're enjoying it. It's been such a passion project, and I am delighted to share it with you all. I hope you don't mind if I copy-paste some of these responses. When they are questions I answer often, I would rather give one quality response copied multiple times, rather than incomplete answers written by hand each time.
Q1. Copied from an email to a beta tester: "Unfortunately, the vertical speed reactivity does really seem to be at a local maxima due to the canard aerodynamics. I spent a lot of time trying to get some of the functionality that might help alleviate the problem working, but I only confirmed for myself that those settings don't work, despite being documented in MSFS 2020 and 2024. Since we were debating the severity of the issue, I was testing while watching the accelerations. The most severe pitch change I could induce was by commanding -2,000 to +2,000 fpm, which resulted in +/- 0.5 G's. Definitely more than would be comfortable, but at least the items in the cabin are staying on the floor. More reasonable vertical speed changes, like 0 to +1,500 fpm resulted in 0.35 G's."
TL;DR: The autopilot's pitch control needs to be "tight" to maintain stability with MSFS's very limited ability to simulate canards. Any less aggressive, and the aircraft will enter oscillations at very high altitudes. There are some features in MSFS that are supposed to allow me to fine tune this, but they haven't seemed to be working since the release of MSFS.
Q2. Copied from another post yesterday: "This is one of the only autopilot limitations I wasn't able to come up with an elegant solution for. You just have to select a desired altitude first (possibly activate another vertical mode) and then the pitch trim wheel will take over. This only applies if you are coming from ALTS."
Q3: Yes, the Linear/Angular deviation flag has been replaced by the cyan "TERM" flag when the FMS is in the approach phase. The original version of Starship's avionics didn't have an approach database, so this was one of many features added after the fact.
Great questions!
-
Brilliant suggestion to look at what I did in the Duke! I now see why it wasn't working in Starship, and I was able to make the necessary changes. You still have to select a new target altitude, but then adjusting the pitch control wheel will start your climb or descent. Much improved! Thank you so much!
-
@Black-Square You mentioned oscillations at high altitudes, I am currently cruising at 25000ft and fly with FMS 1 as nav source (AP NAV SRC is FMS as well) and NAV Hold selected and I get roll ocillations and my heading changes +-3deg over a period of pretty much 30sec. Is there anything I may be doing wrong?
