Well that’s bizarre. Any nav hold for me ends up with the aircraft constantly turning left and right along the course.
same here, every GA either stock or wt or pms50’s neither works besides heading mode, most just flying parallel, not reacting at all or just straight into the ground.
I had a strange experience last night in the 747. The autopilot was seeing a different North orientation to that on the EFIS and compass. Both in course mode and heading mode. Really really strange. I tested the heading mode and selected due North and it then turned to a heading of 220 degrees. When I then enabled the course mode to follow the flight plan it basically flew a course with a similar deviation in degrees, not following the flight plan at all. There is something seriously not right since SU5. The other thing is that when you select an altitude on the autopilot it never corresponds to the displayed altitude on the EFIS. It’s like it is following some arbitrary pressure setting. I tested this further by setting ICAO standard atmospheric pressure in weather settings and noted at 10000ft selected on the autopilot that it was actually flying at 10600ft.
I’ve had a very frustrating experience yesterday trying to fly an ILS approach using the AP. I could replicate it with the 208 and G36, with all three G1000 configurations (Stock, WT0.4.0, Nxi). It just didn’t want to follow the track precisely. After pressing the APR button and capturing the GS, the aircraft pitched down as expected to follow the GS. It was also trying to follow the LOC as expected. BUT the precision of the AP in following GS/LOC was horrible. The aircraft was not on the GS, but tended to fly either too high or too low. Same with the LOC, it was weaving between slightly left and right of the LOC, but never exactly in the middle. Final approaches are currently only possible to fly by hand, as the AP is incapable of following the ILS with the necessary precision. Keeping the AP activated down until flare will result in a landing anywhere but not on the runway.
Which aircraft are you talking about?
Not a single default one in MSFS is capable of autoland, hence you need to disenage the autopilot between 100-200ft AGL.
2 Likes
G36 and 208. At 100-200ft I’m about ±50m right or left of the runway and a few hundred meters short of the Threshold. It’s so far off, I’d have to disengage it at about 1000ft or it would be impossible to get to the runway. There’s no way the AP can get it down anywhere near minimums safely. I’m 100% sure this is a regression from SU4 to SU5, as it was much more accurate before.
Do you have a few airport + rwy examples where this happens?
I tried EDLN ILS 13 and EHBK ILS 21. Tried it multiple times, the offset is not consistent, the AP will just guide me to a random location somewhere right or left of the runway, mostly far below GS. The ILS itself is perfectly accurate, when hand flying the approach it guides me exactly towards the centerline and threshold. It’s just that the AP is incapable of following the ILS within any reasonable accuracy, so the APR mode is basically useless.
It’s important to note that the G1000 stock and 0.4.0 MFDs will always show your own-aircraft symbol not completely overlaid on the Magenta line.
Zoom in or out and you will see that overlay occur. It’s simply a visual glitch, the AP is still following NAV guidance.
Folks - the bottom line is this - if you are using a NavUnit that has a Working Title equivalent (Rockwell Colline Proline 21, G1000, G3000, G3X), you should be using the WT enhancement. This is what it’s going to be in the future when those mods get stabilized and become part of core sim code.
To a lesser extent, there are enhancements to the GNS 430/530 by PMS, but that’s a true Third Party, so we can’t speak to it in the same way that Working Title is underwritten and endorsed by Asobo to become part of core sim code. Nonetheless, PMS is an improvement over stock GNS.
The jetliners are an entirely different story. For anyone who’s used FBW’s A320 those folks also write their own flight planning logic overlaying the Asobo stock FMS, similar to Working Title. There are way too many nuances there and I’m not close enough to liners to speak more than at this generality level.
It’s important to note that the inaccurate path following of the AP on Approach (maybe also during flight as reported by others) happens with all of the G1000 configurations. Stock, 0.4.0 and Nxi. Added this information to my first post. I’m sure this is not just a visual glitch, as the airplane is physically weaving left and right from the track (LOC), and it’s more than just a bit. Something happened to the accuracy of the AP with SU5.
I’m not seeing this on NXi. The path is not weaving if I zoom down. To be clear, is this happening on the MFD or the pseudo VFR Moving Map window?
I think on all of them, but I primarily look at the HSI and the runway. And both the green LOC line on the HSI and my eyes tell me that I’m not aligned with the runway at all when using the AP. I’ll try to make a video later today so it will be more clear.
The CDI is not auto-slewing on ILS guidance. AP is still guiding, but unless you used the CRS knob to manually align the CDI with the runway’s magnetic bearing, it will look “off.” That was enabled in WT G1000 0.3.7, but is pending on the NXi. Please see Matt/WT (Bishop398) comments on the G1000 NXi Master Thread regarding this.
Just tested EHBK ILS 21 with the 172 ‘classic’ and you are correct, the AP ILS tracking is horrible!
1 Like
I’ve also shortcut ILS approaches by DTO to the FACF and switching the CDI manually to the tuned NAV radio. It looked and guided as expected. LOC guidance correction aligned with expected runway “visual” confirmation, followed by GS capture just after the FAF.
Even if the loc ccourse is set wrong, the AP should track the LOC correctly since the LOC deviation is
correctly displayed, regardless of the set course.
1 Like
LOC and GS capture works nicely, but the closer to the runway you get, the greater the LOC & GS deviations.
I think that varies by airport though? The last couple of approaches I shot were right on - double-checking with the plates.
I’ve tried the EHBK 21 approach three times (calm weather) and the AP ILS tracking differed each time.
Worth reporting to WT on the NXi thread (this was NXi?).