Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Aircraft gives too large inputs when letting go of the stick #9399

Open
Jaronics opened this issue Nov 4, 2024 · 60 comments
Open

Aircraft gives too large inputs when letting go of the stick #9399

Jaronics opened this issue Nov 4, 2024 · 60 comments
Labels
A380X Related to the A380X aircraft Bug Something isn't working Flight Model Anything thats related to the planes performance regarding aerodynamics etc. Reproduced Bug has been verified and reproduced by QA

Comments

@Jaronics
Copy link

Jaronics commented Nov 4, 2024

Aircraft Version

Stable

Build info

{
    "built": "2024-11-02T19:03:13+00:00",
    "ref": "refs/tags/v0.12.1",
    "sha": "2776100c286d3723826c62df36dc48207657d937",
    "actor": "Benjozork",
    "event_name": "manual",
    "pretty_release_name": "stable/v0.12.1",
    "version": "a380x-v0.12.1-rel.2776100"
}

Describe the bug

When moving the stick and letting go, the plane rolls back a bit.

Expected behavior

The plane stabilizes after letting go of the stick and doesn't roll back.

Steps to reproduce

Takeoff

Roll the plane to one side and let go of the stick

Observe the roll indicator

References (optional)

https://youtu.be/PtDehEfu57Q Video of me testing the roll.

Additional info (optional)

No response

Discord Username (optional)

No response
@Jaronics Jaronics added A380X Related to the A380X aircraft Bug Something isn't working labels Nov 4, 2024
@donstim
Copy link
Contributor

donstim commented Nov 5, 2024

Another example here: https://youtu.be/DnY5YY5cCyE?si=OnFMJMdd2IbpqC6v&t=3762

@Jaronics
Copy link
Author

Jaronics commented Nov 5, 2024

Yes, that's it.

@AlonsoJCY
Copy link

Yep, I think that is one of the major problem that make the landing difficult to keep stable

@alepouna alepouna added Flight Model Anything thats related to the planes performance regarding aerodynamics etc. Reproduced Bug has been verified and reproduced by QA labels Nov 9, 2024
@alepouna
Copy link
Member

alepouna commented Nov 9, 2024

QA Reproduced

@colinlenzner
Copy link

Have the same problem. You can see it very nicely on the F/CTL page. After „finishing“ my input the aircraft makes an input on its own into the other direction. I habe the feeling that the problem gets worse when on approach and fully configurated. Makes the plane pretty much unflyabke for me, sadly, as especially the landing is pretty much a fight of me against the aircraft.
I‘m using the Thrustmaster Airbus TCA sidestick, if thats of any relevance.

@exocetfalling
Copy link

I have this too with a VKB Gladiator NXT.

@admccarty
Copy link

Same here. I have tried on both the T.flight Hotas X and the Thrustmaster Airbus TCA and both have the same issue. Literally cannot land unless it's left on autopilot, completely unflyable on final approach when manually flying as the plane fights back in the opposite direction to where you're telling it to go.

@FWB800
Copy link

FWB800 commented Nov 13, 2024

I always find problems like this difficult to analyse, I'm not having the pitch and roll issues that are being reported, I don't know why I don't. I'm using TCA Airbus Captains X pack on a pc I bought in May 2024. If I do a slight roll and leave go, it corrects to level flight and doesn't go beyond that and because of that I find it very stable on finals. If I do the max bank and leave go, it goes back to a stable 30 deg turn on it's own and stays there, pitch control is good and doesn't appear to be over sensitive, using same config for the A32NX on the A380X for the sidestick

@talkedeh96

This comment was marked as duplicate.

Sign in to view
@colinlenzner
Copy link

Just wanted to add that the problem seemingly worsens with the flaps extended.

@Jaronics

This comment was marked as duplicate.

Sign in to view
@exocetfalling
Copy link

Any news on this? I have tried to solve it myself but no luck so far.

I've tried to adjust things in [flight_tuning] but increasing aileron effectiveness just makes it worse. The issue probably lies in the MATLAB FBW laws code, so unless we can fix that we're stuck.

@colinlenzner
Copy link

Any news on this? I have tried to solve it myself but no luck so far.

I've tried to adjust things in [flight_tuning] but increasing aileron effectiveness just makes it worse. The issue probably lies in the MATLAB FBW laws code, so unless we can fix that we're stuck.

But alepouna was able to reproduce the issue. So shouldn't it be able to already be trackable? I mean, some users don't have it apparantly. So it can't be a general problem for everyone, right?

@Jaronics
Copy link
Author

Jaronics commented Dec 1, 2024

I have got a new stick now and the problem still persists. Although it seems to be less. I have only done one flight so far, so I will need to test further.

@leftseat380
Copy link

I can confirm this as well. When able, I will check my flight controls as well when rolling. Does the rudder move when leveling the wings?

@admccarty
Copy link

Anyone with this issue tested the A380 in MSFS 2024? Would be interesting to see if the issue persists in the new sim.

@colinlenzner

This comment was marked as spam.

Sign in to view
@exocetfalling
Copy link

exocetfalling commented Dec 20, 2024 via email

@Jaronics
Copy link
Author

There is work going on. With #9645 the issue seems to be fixed.

Any updates???

@donstim donstim mentioned this issue Dec 31, 2024
Closed
@Jaronics
Copy link
Author

With the new update the issue seems to be better, but it is still there sometimes.

@donstim
Copy link
Contributor

donstim commented Dec 31, 2024

Thanks for the report.

@colinlenzner
Copy link

Yes I‘d also say it has gotten better. But it still is present. I‘d say the slower you let go of the stick/return to neutral the less pronounced the issue is? But also during landing I feel its more prominent than on takeoff.

@colinlenzner
Copy link

So I've just done a few complete flights with the aircraft. I can definitely say it has gotten better, as stated above. BUT especially on approach its as bad as before to my taste. To counter this you basically have to give some kind of input all the time it feels like, so that you do not return the stick to the center quickly in any way. This results in constant slow inputs. I've just done a landing in gusting winds up to 40kts and with this one the issue was really really noticeable again. On the other flights with calm winds it isn't that big of an issue as you don't have to correct that much.
What gives me hope though is that one of the earlier dev-patches did make it in fact better. IIRC it fixed some other flight dynamic stuff. What also gives me hope is that @alepouna mentioned the issue is reproducable. Although I'd love to hear what exactly the cause is (as apparently it is known).
I've also played around with my joystick settings again, creating huge neutral deadzones and turning the reactivity down and whatever but nothing changes anything. Besides that my joystick is also not noisy or anything and I've tested the aircraft with two joysticks and a yoke in the meantime.
Last thing: I've also noticed a kind of similar issue on the ground with the nosewheel steering? I use my twist axis on the joystick for the tiller input and I noticed when letting go of it quickly the aircraft also steers a little bit into the other direction. But I have no clue if that is supposed to be like that, that it is some kind of tirewall flexing or something like that? Never had it on any other airplane though.

@Jaronics
Copy link
Author

Jaronics commented Jan 8, 2025

I noticed the thing with the nose wheel as well. I am using the Z axis of my stick to steer on the ground.

@colinlenzner
Copy link

I noticed the thing with the nose wheel as well. I am using the Z axis of my stick to steer on the ground.

Same, using the twist axis. But I've played around so much with sensitivity settings, checked the axes in MSFS and also straight in Windows and there is 1000000% definitely nothing that corresponds to the aircrafts' behaviour.

@admccarty
Copy link

Haven't flown this plane in a while but came back to test it again after seeing these comments.

Still seeing the same issue here with the latest dev update installed, doesn't matter if roll is set as linear or not I get the same results. I think it is slightly less severe than the last time I flew the A380, but definitely still there. Wind was totally calm when I tested.

@donstim
Copy link
Contributor

donstim commented Jan 21, 2025

Interesting that you all are still experiencing this issue. Can you post a video showing your sidestick movement and PFD? See my video above. Here are 2 more, one at 5000 ft at 250 KCAS flaps up, and one on a nominal approach path and speed with full landing flaps. If anything, for larger stick movements, I am seeing a bit of continued roll for a bit after the stick is returned to neutral, though I don't know how fast the airplane can stop the roll IRL. I am not seeing any rollback in the other direction at all.

https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxTfSWE9Mfk3mzUyA?e=3vB5JZ

https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxUqMFYVC4kR5S5SQ?e=5jeZaA

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@donstim
Copy link
Contributor

donstim commented Jan 21, 2025

Again, I invite you to provide a video of such, showing both the sidestick (or controller input) and PFD (or other indication of bank angle/roll rate) as well as pertinent parameters like weight, speed, CG, airplane configuration and flight path. Anything that would help me to reproduce this issue.. I've provided 3 videos showing that I cannot reproduce it. If it can't be reliably reproduced, it can't be fixed.

@exocetfalling
Copy link

exocetfalling commented Jan 21, 2025 via email

@exocetfalling
Copy link

exocetfalling commented Jan 21, 2025 via email

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@donstim
Copy link
Contributor

donstim commented Jan 21, 2025

https://youtu.be/3oq52QQPYX8

See this video, taken today with version 0a7533e.
…

Thanks for this. That's definitely strange. Do you also see this behavior on our A32NX?

What type of controller do you have? If you test the controller in the Windows controller properties page, does it overshoot when you release to neutral? How about if you do that with the controller sensitivity window in MSFS open?

@donstim
Copy link
Contributor

donstim commented Jan 21, 2025

even on the ground I get opposite yaw.I use my aileron axis for my nw
steering
and When I go to neutral it turns back slightly.
…

That just really sounds like a controller issue if it happens in both the yaw and roll axes. I don't think steering on the ground has anything to do with roll response in flight.

@exocetfalling
Copy link

exocetfalling commented Jan 21, 2025 via email •
edited
Loading

@exocetfalling
Copy link

exocetfalling commented Jan 21, 2025 via email

@Jaronics
Copy link
Author

Yes, it doesn't happen on the a32nx.

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@Jaronics
Copy link
Author

Thats pretty hard reaction tho. For me it is way less with the new versions of the plane

@exocetfalling
Copy link

exocetfalling commented Jan 21, 2025 via email •
edited
Loading

@leftseat380
Copy link

leftseat380 commented Jan 21, 2025 via email

@donstim
Copy link
Contributor

donstim commented Jan 22, 2025

https://youtu.be/5V38hi49T0E

A32NX today, no issue. You can see the roll is crisp as can be.
…

What I see is the exact same issue. Look at your first two inputs. Right after you neutralize it comes right back to the original back angle, just like in your A380X video. I don't think this is the behavior that the others are experiencing. This is well beyond the ordinary. I suspect an issue with your controller.

This is what I get for the A32NX roll response: https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxav_CKeV1X1vATAg?e=SJTh8Z

Here is what I get for the A380X: https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxbK2-ohSE8B5Q6sw?e=yuhKAK

I can now see that for the slightest of inputs, it does appear to roll back a bit. Is this the behavior you are seeing, for example, @Jaronics? For most inputs I don't see any of that at all, but for the slightest ones it does show up for me, though it is not noticeable to me when flying the airplane. If that's it, I'll have to see what, if anything can be done about it.

@exocetfalling
Copy link

exocetfalling commented Jan 22, 2025 •
edited
Loading

It's probably not the controller because, if it were, the side stick would reverse direction too. But it doesn't, it's only the aircraft that does that. Now that you mention it, I can see a bit of the issue on the A32NX but nowhere enough to be a problem. So something is different between the A32NX and A380X.

@talkedeh96
Copy link

Here is a clip of the nose wheel steering issue discussed: https://streamable.com/kypho2

It is very noticeable how when you release it will go in the opposite direction then recenter. This is the same issue with the aileron movements reported.

@BlueberryKing
Copy link
Member

Here is a clip of the nose wheel steering issue discussed: https://streamable.com/kypho2

It is very noticeable how when you release it will go in the opposite direction then recenter. This is the same issue with the aileron movements reported.

That is a different issue. Check that you have Auto-rudder disabled in your assistance settings. We can also provide further support on Discord.

@leftseat380
Copy link

leftseat380 commented Jan 22, 2025 via email

@talkedeh96
Copy link

Here is a clip of the nose wheel steering issue discussed: https://streamable.com/kypho2
It is very noticeable how when you release it will go in the opposite direction then recenter. This is the same issue with the aileron movements reported.

That is a different issue. Check that you have Auto-rudder disabled in your assistance settings. We can also provide further support on Discord.

I do have Auto-rudder and all assisted disabled, I also posted in discord and received the same answer. I think the issues are related as in both bugs, the plane goes in the opposite direction upon centering.

@77W9940 77W9940 mentioned this issue Feb 7, 2025
Merged
@Jaronics
Copy link
Author

https://youtu.be/5V38hi49T0E
A32NX today, no issue. You can see the roll is crisp as can be.
…

What I see is the exact same issue. Look at your first two inputs. Right after you neutralize it comes right back to the original back angle, just like in your A380X video. I don't think this is the behavior that the others are experiencing. This is well beyond the ordinary. I suspect an issue with your controller.

This is what I get for the A32NX roll response: https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxav_CKeV1X1vATAg?e=SJTh8Z

Here is what I get for the A380X: https://1drv.ms/v/s!Ajsy7Ey4_8uFgZxbK2-ohSE8B5Q6sw?e=yuhKAK

I can now see that for the slightest of inputs, it does appear to roll back a bit. Is this the behavior you are seeing, for example, @Jaronics? For most inputs I don't see any of that at all, but for the slightest ones it does show up for me, though it is not noticeable to me when flying the airplane. If that's it, I'll have to see what, if anything can be done about it.

I have a new theory on this. For me, it seems like it depends on performance. Just now I had a really laggy sim and the plane was nearly uncontrollable. I have had landings where it was better and the performance of the game was also better. I dont know if that is possible but that is how it looks to me

@Jaronics
Copy link
Author

Same with the rudder and tiller issue. I think it has to do with performance. I can't explain otherwise why it is sometimes so strong and sometimes not even there.

@aviomare
Copy link

I spent a lot of time trying to reproduce this issue and it appears that there are different root causes, however this one is the most common for me. If you introduce an aileron input, but release the sidestick before the aircraft starts banking, it's going to bank into the opposite direction. Basically, something like this:

  • Introduce some smaller left aileron input
  • Nothing happens for about 2 seconds
  • Release the sidestick (aileron)
  • Aircraft starts violently banking to the right side

It almost feels like there's a delay when you input aileron (so that aileron is not instant, the delay depends on the current bank angle) and if you release the sidestick just when that delay is about to end, it's going to register and perform the opposite movement. In other words, if you do left aileron and nothing happens, you release the sidestick, it registers the sidestick going from the left to the right (i.e. from the left to the neutral), and as a result starts banking the plane to the right side (instead to the left side because that was the original input).

I have a very short video of that happening, sadly you can't see anything else but PFD, but maybe it helps: https://streamable.com/n17zgu

In this video, I introduced the left aileron input, you can see at 0:05 that the plane started banking to the left very little, but then suddenly started turning right even though I didn't input any right aileron input. The plane configuration at that flight was:

  • PAX 428
  • Cargo 8560 kgs
  • ZFW 344519 kgs
  • ZFWCG 35.27%
  • Fuel 103500 kg

I originally posted this on Discord on 2025-02-04, and I used the latest development version available that day.

I saw another example of this issue today on a livestream, but I'm not sure about roll settings and the aircraft version, however I'm providing it as a reference: https://www.youtube.com/live/vUJ1BUtDbGE?t=9717 (the video should start at 2:41:57).

@leftseat380
Copy link

leftseat380 commented Feb 25, 2025 via email

@donstim
Copy link
Contributor

donstim commented Feb 26, 2025

@aviomare Yeah, a video of only the PFD/ND doesn't tell us much. I would like to see at least three other things: (1) Your roll control sensitivity curve, 2) the control inputs being made, and 3) your frame rate using the FPS counter in MSFS dev mode (under the Debug menu). In your instructions, you say that a small aileron input did not produce any response when being held for approximately 2 seconds. That tells me that you either were in or near the deadband of your controller sensitivity curve, or your frame rate was low.

Please use this controller display (or something similar) to show your control inputs:
Joystick-Display.zip
(You just need to unzip it, run the .exe, and then right click in the display and select "connect."

Here is the sensitivity curve you should start with, only modifying if necessary for your controller to get the right control response:

Image

Here is my attempt at reproducing what you described. You can see there is very little response delay and no rollback in the other direction: https://1drv.ms/v/s!Ajsy7Ey4_8uFgZ09P4Gx5xjfmX9oqA?e=WSXPJb I regret not showing the FPS counter.

Not that it matters, but judging from the speeds shown in the PFD in your video, you were lighter than what you stated as your airplane configuration at the time. The weight you gave is far above the max landing weight of the airplane.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
A380X Related to the A380X aircraft Bug Something isn't working Flight Model Anything thats related to the planes performance regarding aerodynamics etc. Reproduced Bug has been verified and reproduced by QA
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests