Results 1 to 15 of 15

Thread: Flight model rudder

  1. #1
    Veteran Combat pilot
    Join Date
    Mar 2013
    Posts
    326
    Post Thanks / Like
    Total Downloaded
    589.91 MB

    Flight model rudder

    Can we raise a bug fix request for the rudder part of the flight model please? Specifically, the nodding from side to side when more than half or so rudder input is applied.

    It is not realistic and prevents manoevres such as stall turns, side-slipping and flick rolls (ie high-speed, accelerated stall, 'spins' in the direction of flight) being properly performed.

    There is a thread over on the Il2Sturmovik forum where someone has looked at this in some detail. One interesting discovery they seem to have made is that when full rudder is applied to recover from a spin, it is possibly hard-coded that a set number of seconds has to run before the recovery takes effect. The result is that whereas standard spin recovery is not seemingly affected (as IRL it takes a second or two for the full recovery rudder input to take effect at the low airspeed prevailing), on the other hand recovery from a flick roll - which (due to the much higher airspeed) should be instantaneous as soon as the wing is unloaded by centralising the stick - also incorrectly takes several seconds during which the nose nods all over the place.

    It could be that the coding for the rudder is overly complex. All it needs is for the primary yaw effect to progressively increase as rudder input is increasingly applied, and the secondary effect of roll in the direction of yaw to do the same.

    If this could be fixed then we'd probably see fewer take-off prangs besides the increased realism!

    Thanks!

  2. Likes von Graf, N/A liked this post
  3. #2
    ATAG Member ATAG_kiwiflieger's Avatar
    Join Date
    Jan 2017
    Location
    Top of the South, New Zealand
    Posts
    664
    Post Thanks / Like
    Total Downloaded
    121.95 MB

    Re: Flight model rudder

    This is something I've been thinking about for a while too, all the aircraft in Cliffs seem to have really poor directional stability. I can't make any comments on how close that is to the real aircraft but it does seem to be overdone, flight experience in gliders shows that even without several hundred HP up front to pull the nose straight, you still have good damping and if you do apply rudder the nose will return pretty much directly to straight into the oncoming airflow without wandering around and oscillating past neutral.

    I just ran some tests in the 109E and I'm not convinced a coded delay in the recovery time exists, although flick rolls do seem to 'bite' in and unless you apply corrective control immediately after the entry you'll do more than one roll. I also noticed that the pitch tended to oscillate with some recoveries so maybe the poor stability also extends to pitch?


    "Speed is life. Altitude is life insurance. Surprise is key."


    ~ AMD FX-4100 3.6GHz (OC to 4GHz) ~ 8GB RAM ~ ASUS GTX1050Ti Phoenix 4GB ~
    T.16000M ~ TWCS Throttle ~ TFRP Pedals ~ DIY IR Headtracker


  4. Likes von Graf, carbolicus liked this post
  5. #3
    Supporting Member
    Join Date
    Aug 2020
    Location
    Germany, Ruhrpott
    Posts
    60
    Post Thanks / Like
    Total Downloaded
    153.35 MB

    Re: Flight model rudder

    I don't really play CloD for this reason. Tried some 'workarounds', like extreme settings, external software, etc. but didn't get good results. If at all.

  6. #4
    Combat pilot Little Bill's Avatar
    Join Date
    Apr 2018
    Location
    West Coast, USA
    Posts
    191
    Post Thanks / Like
    Total Downloaded
    465.46 MB

    Re: Flight model rudder

    . carbolicus I like your post. In regards to calling this a bug, I have a question as there are really may be two issues here with the programming. One is the response to rudder input in what I will call normal flight maneuvers (the walling around of the noise) and the other is in recovery from a spin. My question is do we know what real 1940ish fighter aircraft did? And more specifically do we know what each aircraft modeled in Cliffs did back then? If we do and their behavior is not what the computer is doing for us now then there should be a bug or 2 called out, if we are only going from what other simulations do then I question the idea. But as a onetime real pilot I agree, rudder response at some airspeeds should be looked at.

    I7 12700K @ 4.8Ghz|RTX 3080Ti|32GB DDR4|1TB NvMe l LG OLED 48”TV|1440P @120Hz|VKB STECS l VKB Gunfighter III/100mm extension l MFG V3 Pedals l Falcon Gear Quadrant l TIR 5 l Cloud X Headphone|Widows 11, C:\Program Files (x86)Steam\steamapps\common\IL-2 Sturmovik Cliffs of Dover Blitz|20Mbs dl, 2Mbs ul

  7. Likes carbolicus liked this post
  8. #5
    Supporting Member 9./JG52_J-HAT's Avatar
    Join Date
    May 2012
    Location
    Germany
    Posts
    1,367
    Post Thanks / Like
    Blog Entries
    16
    Total Downloaded
    266.37 MB

    Re: Flight model rudder

    https://forum.il2sturmovik.com/topic...y-in-this-sim/

    Maybe the discussion there can be taken as additional input?
    i7 13700KF @5.3 Ghz | RTX 4090 526.86 1440p | 32 Gb RAM 6000 | Win10 64 bit | TrackIR 5 | VKB Gunfighter III | MFG Crosswind V2 | Warthog Throttle | Reverb G2 100% SteamVR SS

  9. Likes carbolicus, von Graf liked this post
  10. #6
    Veteran Combat pilot
    Join Date
    Mar 2013
    Posts
    326
    Post Thanks / Like
    Total Downloaded
    589.91 MB

    Re: Flight model rudder

    Quote Originally Posted by Little Bill View Post
    . carbolicus I like your post. In regards to calling this a bug, I have a question as there are really may be two issues here with the programming. One is the response to rudder input in what I will call normal flight maneuvers (the walling around of the noise) and the other is in recovery from a spin. My question is do we know what real 1940ish fighter aircraft did? And more specifically do we know what each aircraft modeled in Cliffs did back then? If we do and their behavior is not what the computer is doing for us now then there should be a bug or 2 called out, if we are only going from what other simulations do then I question the idea. But as a onetime real pilot I agree, rudder response at some airspeeds should be looked at.
    I obviously have no experience of flying a Spitfire etc - if only, it would be a dream come true!
    But from what we can read in books etc about how these old aeroplanes flew, whether Spits, Hurricanes or Bf109s, they do seem to have been well-mannered and safe aircraft to fly with conventional spin recovery characteristics, so perhaps if they enter and recover from spins in a similar manner to, say, a Cessna 152, then that would be quite acceptable.

    What I am quite certain about is that no conventional aeroplane which is certified safe to fly would respond to rudder inputs like all CLoD aircraft do with that manic nose-nodding (or tail-wagging which might be a better way of putting it).

    I would imagine that if a real aeroplane were to behave like that then it would be condemned as unsafe due to the fin being too small and not damping yaw sufficiently.

    I certainly think that all the CLoD fighters are dangerous to the pilot when taking off as you have to do such a complicated dance on the rudder pedals to make it safely into the air. Most people manage it OK because we do it a lot and get used to it, but I do have a friend (a retired guy) who loves this sim but just can't do take-offs, so he always has to start in the air. Hence he can't fly full-real on-line - and finds it hard to believe me when I tell him that real aircraft - even tail-draggers - are actually far easier to take off!

    So I guess this isn't a software bug as such, more a failing in systems analysis in that the developers did not understand properly how a rudder works, and presumably a real-life pilot was not involved in testing.

  11. Likes N/A, von Graf liked this post
  12. #7
    Team Fusion ATAG_Noofy's Avatar
    Join Date
    Feb 2019
    Location
    https://w3w.co/chat.hisses.lofty
    Posts
    2,602
    Post Thanks / Like
    Total Downloaded
    1.25 GB

    Re: Flight model rudder

    If you put too much gas from the start, the engine torque will prevail.
    Usually planes want to go left because of the torque and wash, so just keep slight right rudder until you build up some speed and rudder becomes efficient.
    Basically when you can start to push the nose down (gently) to lift the tail, you know you’re fine.
    The error many pilots do in the beginning is to try to over correct.
    Just be gentle with these planes, especially those with a narrow landing gear (spits and Bf109s).
    This is the key. Be gentle.
    Gigabyte Z390 UD | i7 9700K @3.60GHz | 16.0 GB | Windows 10 Pro 64-Bit | NVIDIA GeForce GTX 1070 Ti
    TM Warthog HOTAS | Saitek pro rudder pedals | TrackIR 5 | TeamSpeak 3.3.2 | TS Notifier 1.6.0h

  13. #8
    Veteran Combat pilot
    Join Date
    Mar 2013
    Posts
    326
    Post Thanks / Like
    Total Downloaded
    589.91 MB

    Re: Flight model rudder

    Quote Originally Posted by ATAG_Noofy View Post
    If you put too much gas from the start, the engine torque will prevail.
    Usually planes want to go left because of the torque and wash, so just keep slight right rudder until you build up some speed and rudder becomes efficient.
    Basically when you can start to push the nose down (gently) to lift the tail, you know you’re fine.
    The error many pilots do in the beginning is to try to over correct.
    Just be gentle with these planes, especially those with a narrow landing gear (spits and Bf109s).
    This is the key. Be gentle.
    Agree with you 100% Noofy, that's the way to take-off a high-powered tail-dragger, and it helps with CLoD fighters too, but for a different (and un-real) reason.

    To summarise, IRL on take-off for many fighters, you used less than full power at first because the rudder with little airflow over it can't fully stop the torque swing. Once the tail is up and more rudder authority available you can add full power and still control the swing.

    In CLoD you also need less than full power at first - but in CLoD's case it's to avoid the need for more than half rudder, hence avoiding the unrealistic (and prang-inducing) nose-nodding (which easily becomes uncontrollable)!

    So I'm afraid the CLoD rudder flight model is still wrong, and makes taking-off more difficult than it should be even for a 1000 hp fighter.

  14. #9
    Team Fusion ATAG_Noofy's Avatar
    Join Date
    Feb 2019
    Location
    https://w3w.co/chat.hisses.lofty
    Posts
    2,602
    Post Thanks / Like
    Total Downloaded
    1.25 GB

    Re: Flight model rudder

    Quote Originally Posted by carbolicus View Post
    In CLoD you also need less than full power at first - but in CLoD's case it's to avoid the need for more than half rudder, hence avoiding the unrealistic (and prang-inducing) nose-nodding (which easily becomes uncontrollable)!

    So I'm afraid the CLoD rudder flight model is still wrong, and makes taking-off more difficult than it should be even for a 1000 hp fighter.
    The difference with RL may be the lack of physical sensations. You do not feel the plane movements, the shaking of the stick, etc. However with some experience this becomes almost perceivable... Otherwise the flight modeling is impressively accurate I must say.

    I am not sure I quite understand what you mean by "nose-nodding". Is this left-right or up-down? Try adjusting your axis sensitivities and/or dead zones.
    Gigabyte Z390 UD | i7 9700K @3.60GHz | 16.0 GB | Windows 10 Pro 64-Bit | NVIDIA GeForce GTX 1070 Ti
    TM Warthog HOTAS | Saitek pro rudder pedals | TrackIR 5 | TeamSpeak 3.3.2 | TS Notifier 1.6.0h

  15. #10
    ATAG Member ATAG_Flare's Avatar
    Join Date
    Oct 2013
    Location
    Interior BC --> Kingston ON
    Posts
    2,801
    Post Thanks / Like
    Total Downloaded
    383.91 MB

    Re: Flight model rudder

    I think he means the under-damped nature of the oscillations you get after applying yaw. I just went and did a test and entered into flick rolls in the 109F, Spit V, Hurricane, Gladiator, and Su-26.

    I was able to recover from the flick roll after 1 roll in the 109 sometimes, but usually needed to let it roll around a second time. The Hurricane and Spitfire needed at least 2-3 rolls to exit the flick roll. The gladiator I had a hard time even entering an accelerated stall.

    In the Su-26 flick rolls can be performed by the book and it will exit from the flick roll exactly as you would expect.


    When I say flick roll I mean pulling back on the stick to cause a stall, then applying hard rudder, no aileron. The flick roll should stop when the rudder is returned to centre and stick pushed forward.

  16. Likes von Graf liked this post
  17. #11
    varrattu
    Guest

    Re: Flight model rudder

    Hello all,

    enclosed please find a snippet from handling and manoeuvrability tests on a captured BF109. Maybe it helps a little for further conversation. The 1942 handling test covered ease of take-off and landing ; trim and stability ; " one control " tests, flat turns and sideslips ; stalling tests, including a determination of Cl max ; high-speed dive ; harmony and " feel " of the controls.

    ~V~



    Quote Originally Posted by carbolicus View Post
    ... from what we can read in books etc about how these old aeroplanes flew, whether Spits, Hurricanes or Bf109s, they do seem to have been well-mannered and safe aircraft to fly with conventional spin recovery characteristics, so perhaps if they enter and recover from spins in a similar manner to, say, a Cessna 152, then that would be quite acceptable.
    Attached Images Attached Images

  18. Likes von Graf liked this post
  19. #12
    Veteran Combat pilot
    Join Date
    Mar 2013
    Posts
    326
    Post Thanks / Like
    Total Downloaded
    589.91 MB

    Re: Flight model rudder

    Quote Originally Posted by ATAG_Noofy View Post
    The difference with RL may be the lack of physical sensations. You do not feel the plane movements, the shaking of the stick, etc. However with some experience this becomes almost perceivable... Otherwise the flight modeling is impressively accurate I must say.

    I am not sure I quite understand what you mean by "nose-nodding". Is this left-right or up-down? Try adjusting your axis sensitivities and/or dead zones.
    Hi Noofy - Just fly straight and level, then apply increasing amounts of rudder. After about half rudder the nose starts to swing left/right - ie yaw is seemingly undamped as if the fin has suddenly halved in size!

    No aeroplane I have ever flown in real life (including vintage tail-draggers, aerobatic types and multi-engine) has ever behaved like that, and I wouldn't fly it again if it did!

    I do agree with you though that the rest of the flight model is very good.

    I suspect that most people do not notice the idiosyncratic rudder behaviour because no-one in combat is going to fool around with stall turns and there is never any need to side-slip to a landing. And if you do get into a spin it will recover using the standard procedure.

  20. #13
    Veteran Combat pilot
    Join Date
    Mar 2013
    Posts
    326
    Post Thanks / Like
    Total Downloaded
    589.91 MB

    Re: Flight model rudder

    Quote Originally Posted by ATAG_Noofy View Post
    The difference with RL may be the lack of physical sensations. You do not feel the plane movements, the shaking of the stick, etc. However with some experience this becomes almost perceivable... Otherwise the flight modeling is impressively accurate I must say.

    I am not sure I quite understand what you mean by "nose-nodding". Is this left-right or up-down? Try adjusting your axis sensitivities and/or dead zones.
    Noofy - Just taken your advice and have played with rudder sensitivities and dead zones - and have been able to improve matters significantly! I think there is still a damping problem (see other posters' comments) and some swaying around is still there, but getting things into perspective, I'm sure there are more urgent issues for TF to address (new weather and VR for starters!) Thanks for your input.

  21. #14
    Team Fusion ATAG_Noofy's Avatar
    Join Date
    Feb 2019
    Location
    https://w3w.co/chat.hisses.lofty
    Posts
    2,602
    Post Thanks / Like
    Total Downloaded
    1.25 GB

    Re: Flight model rudder

    Quote Originally Posted by carbolicus View Post
    I suspect that most people do not notice the idiosyncratic rudder behaviour because no-one in combat is going to fool around with stall turns and there is never any need to side-slip to a landing. And if you do get into a spin it will recover using the standard procedure.
    I use full rudder extensively for slowing the plane down, often for landing, occasionally in combat. Never noticed a "damping" effect...

    I usually go right rudder with compensating left aileron. Not sure if I ever tried it the opposite way. Could it be different left from right due to prop wash?

    EDIT: Nope, it works fine with me left and right. The key is to do it gradually and to gently compensate with opposite aileron.

    Quote Originally Posted by carbolicus View Post
    Noofy - Just taken your advice and have played with rudder sensitivities and dead zones - and have been able to improve matters significantly! I think there is still a damping problem (see other posters' comments) and some swaying around is still there, but getting things into perspective, I'm sure there are more urgent issues for TF to address (new weather and VR for starters!) Thanks for your input.
    Last edited by ATAG_Noofy; Nov-10-2020 at 17:34.
    Gigabyte Z390 UD | i7 9700K @3.60GHz | 16.0 GB | Windows 10 Pro 64-Bit | NVIDIA GeForce GTX 1070 Ti
    TM Warthog HOTAS | Saitek pro rudder pedals | TrackIR 5 | TeamSpeak 3.3.2 | TS Notifier 1.6.0h

  22. #15
    varrattu
    Guest

    Re: Flight model rudder

    This is what Allied pilots reported about the Bf109 rudder behaviour in 1940 ...

    ~V~

    Attached Images Attached Images

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •