Good evening(+7) When flying a flight plan with real weather and winds of 40-45KT, very often the AP disconnects during the flight specially during climb. The same flight with no weather installed, clear sky, all goes smooth. Is there a bug with rough weather and AP engaged? Now I checked again and the plane is full of ice! This must be the reason (pitot was one at departure). I was flying LFMP-LFBO real weather at 9.000feet!

Thank you for your attention

Jean Paul Reuland

Edited January 9, 20233 yr by Jean Paul Reuland

Well it seems the weather has nothing to do here! I departed from Surabaya with real weather on and 3 min in climb phase, AP disconnected  with only 2KT wind! What could be the reason?

I also engaged FLC during this flight with 90KTS during the climb and could recover the AP flight, Maybe a too slow speed during climbing? Plane just went through a tropical rain and was dancing a Lambada!!! 🙂 But AP stays connected so far. Perhaps it's better to take off with FLC set rather than VS??

JP Reuland

Edited January 9, 20233 yr by Jean Paul Reuland

I don’t know if this is the reason for your issue, but in general it is preferable to climb using FLC to climb at a constant speed rather than a constant rate using VS mode. Several factors could contribute insufficient airspeed using a constant rate climb.

AP will disconnect if for some reason the aircraft stalls. This could be caused by:

.) Setting an excessive vertical speed in VS mode (for a given altitude/weight/thrust setting);

.) Flying in ice conditions, which will cause ice accumulation and an aerodynamic stall;

.) Flying in heavy turbulence, especially at low speed, where some gust could cause an intermittent stall and disconnect the autopilot. X-Plane turbulence seems to be currently overdone in certain conditions, so hopefully it will be improved in subsequent releases.

14 minutes ago, Murmur said:

X-Plane turbulence seems to be currently overdone

I think the tough thing here is "compared to what". 

like we all got used to never really encountering turbulence that had any meaningful impact.

Now it seems a lot more common.

But being used to no turbulence does not mean common turbulence is overdone. 

Most rw plane flights Ive been a passenger on have had at least one "turbulence event" (seat belts on, stuff fly everywhere) flying over trees in a heli on a windy day needs a lot of cyclic and collective input to keep her straight and level.

plus there is currently lots of fairly stormy weather around, US has been suffering terribly aiui, so probably not that surprising to hit somewhat stormy conditions flying state side atm with real weather on?

TLDR: I dont know if it is overdone, "INSUFFICIENT DATA FOR MEANINGFUL ANSWER."

6 minutes ago, mSparks said:

TLDR: I dont know if it is overdone, "INSUFFICIENT DATA FOR MEANINGFUL ANSWER."

Might be, hence I wrote a doubtful "seems". I think recently there have been several YT videos showing turbulence that seemed excessive for the given weather conditions. Let's see what LR decides to do on the matter.

After fifty years of r/w flying (aka MORE THAN SUFFICIENT DATA FOR A MEANINGFUL ANSWER), I can assure you that, yes, as it stands now, airmass instability and its effects are way overdone in XP12 compared to what I've experienced in the real world piloting a wide range of aircraft ranging from sailplanes to 4-engine jet heavies.

We are being told that improvements in the weather model are under very active development--I've taken that at face value and I'm still enjoying P3D and XP11 in the meantime. 

That said, I don't think it's at all productive to deny or minimize problems that clearly do exist at the present time. 

For the OP--another possibility could be a noisy pot (position sensor) in your flight controls making a spurious input that causes an AP disconnect.

Now flying a Cessna 172 with 40 to 45 knots does seem a handful to me? I am not sure irl AP would remain armed, and I am not sure I want to try irl either.

4 minutes ago, soaring_penguin said:

Now flying a Cessna 172 with 40 to 45 knots does seem a handful to me? I am not sure irl AP would remain armed, and I am not sure I want to try irl either.

40-45 knots of wind at altitude is nothing particularly unusual.  As I read the OP, he was in the climb and/or cruising at 9000 ft MSL when he got the disconnects.

 

13 minutes ago, Bob Scott said:

airmass instability and its effects are way overdone in XP12 compared to what I've experienced in the real world piloting a wide range of aircraft ranging from sailplanes to 4-engine jet heavies.

I'm basing what I said on some 35 hours of flight testing long and short haul on the 744 now, at FL150, FL250 and FL390.

The only time I've seen any evidence of the AP responding to air mass stability is passing through a wind layer where the wind changed direction rapidly over a short period of time. Can't say whether that is overdone or not, seems reasonable.

What are you basing that statement on?

Just now, mSparks said:

I'm basing what I said on some 35 hours of flight testing long and short haul on the 744 now, at FL150, FL250 and FL390.

The only time I've seen any evidence of the AP responding to air mass stability is passing through a wind layer where the wind changed direction rapidly over a short period of time. Can't say whether that is overdone or not, seems reasonable.

What are you basing that statement on?

I've been getting very large (e.g. +/- 20 KIAS) fluctuations in airspeed at flight levels in the 30s, not flying near anything convective, and not near a front.  Happened to me repeatedly in different locations and with different acft (CL650, Phenom 300, Citation X) over the course of a week of trials/testing.

I then tried your suggestion to use the instructor panel, and if I use that to set the winds manually with the turbulence dialed down, it's well within the realm of acceptable.  Hence my conclusion that the real weather modelling is the culprit.

30 minutes ago, Bob Scott said:

I've been getting very large (e.g. +/- 20 KIAS) fluctuations in airspeed at flight levels in the 30s,

Ive alsways seen a fair few fluctuations in airspeed + and - 4 or 5 knots, no more, and the new AT handled them much better than the default one did (by ignoring them). again, not sure if thats overdone, seemed reasonable - if I had seen 20kts I would very likely agree, but is it definitely the air changing speed or the AT in the aircraft exaggerating the changes, or the ASI reacting to quickly compared to the real one to said changes?.

There is something wrong with the default yaw damper imho - only Im told that it is the aircraft that need to retune it. Not looked at that for a few months....

30 minutes ago, Bob Scott said:

Hence my conclusion that the real weather modelling is the culprit.

prior to the next/last release (not sure if the latest I was recently testing is out yet) real weather was ctd'ing me a lot, there is also something still funky with what I think is the known issue with cirrus, where high cirrus clouds have walls like pyramids.

My uncertainty comes from what - if anything - is overdone, if their weather source is spitting out that much turbulence in certain locations what can be done about it? If their weather source is spitting out that much turbulence because there is a lot of really weird weather around at the moment (zero snow in europe atm, rage, more than 1 million Americans without power from storm damage, sad face) should anything be done about it?

INSUFFICIENT DATA FOR MEANINGFUL ANSWER.

(the last question, Issac Asimov)

Edited January 9, 20233 yr by mSparks

It wasn't the autothrottles, as the wildly fluctuating wind velocities when coming from a quartering angle also produced a lot of back-and-forth yawing as well.

I suspect that it's not so much an issue of the weather source, but rather how the air mass model is built from the GRIB datasets.  A turbulence attenuation slider similar to what's in the instructor station would be immediately helpful, but beyond that, something is amiss to be producing such exaggerated fluctuations, especially at the upper flight levels.  I've only seen that kind of thing once IRL--in moderate-to-severe mountain wave turbulence in a Gulfstream IV.  I did check the AWC plots and there was no forecast turbulence while this was going on.

One of the real aggravations of not having historical weather capability is not being able to go back and reproduce an issue that occurred at a specific date/time/place.

7 minutes ago, Bob Scott said:

as the wildly fluctuating wind velocities

Here's the thing with that tho:

51 minutes ago, Bob Scott said:

40-45 knots of wind at altitude is nothing particularly unusual.

So any reasonable turbulence there is up to a real +/- 45kts change in velocity over stable air, if one second it is going east, the next second it is heading west....

but  - imho - a real ASI shouldn't capture this.

X-Planes ASI does seem to be capturing this - is it overdone - or is it just showing you what is really happening that you don't normally see, only feel... "don't know"

12 minutes ago, Bob Scott said:

back-and-forth yawing as well.

If you see this, try disabling the yaw damper.. there seems to be an issue with the default YD getting into some kind of reinforcing oscillation state, last time I flew the S76 it was unflyable with the YD on.

4 minutes ago, mSparks said:

Here's the thing with that tho:

So any reasonable turbulence there is up to a real +/- 45kts change in velocity over stable air, if one second it is going east, the next second it is heading west....

but  - imho - a real ASI shouldn't capture this.

X-Planes ASI does seem to be capturing this - is it overdone - or is it just showing you what is really happening that you don't normally see, only feel... "don't know"

Sorry, I'm not making any sense of this.  Steady-state winds of 45 knots at altitude are normal--sudden wind shifts that large are not unless maybe you're caught in a microburst.