Wing snap vs joystick?

[VZ_342]

Just curious at this point, but I suppose we should look into this...

 

Please reply with what joystick or base you use, FFB used or not, and whether or not you experience wing snap.

 

I use a TM Warthog w/F18 stick, and although I can avoid snap at easy conditions, in a high stress environment (combat) snap happens every time.

[MAXsenna]

MS FFB2. FFB displacement setting in the Special Options. I have to try to snap'em off.

Sent from my SM-A536B using Tapatalk

[Bucic]

Xbox controller

Saturation Y 85

Curvature 20

Dead zone 1

I have to try hard to snap the wings off.

Here's a video of the response of the pitch channel to immediate full defections back and forth with no aerodynamic load (end of the video).

https://youtu.be/pJIakZ74ycw

 

[Bremspropeller]

It's not a joystick thing, its more the lack of a proper q-feel system (see F1 and F-4) that makes the wings snap. The joystick is just a mitigation, not eliminating the root cause.

The way the snaps are modelled is a bit overzeleaous, as has been mentioned by a couple of people a couple of times.

 

I'd love to see the accident-reports of multiple F-5s shedding wings because of some rolling gs. If it was that easy, we'd have historical evidence of it.

[Bucic]

16 minutes ago, Bremspropeller said:

It's not a joystick thing, its more the lack of a proper q-feel system (see F1 and F-4) that makes the wings snap. The joystick is just a mitigation, not eliminating the root cause.

The way the snaps are modelled is a bit overzeleaous, as has been mentioned by a couple of people a couple of times.

 

I'd love to see the accident-reports of multiple F-5s shedding wings because of some rolling gs. If it was that easy, we'd have historical evidence of it.

Disregarding the limitations your premise for the real life cases is:

pilots do the training and learn about the airframe structural limitations - pilots disregard the limitations and exceed them daily - nothing happens

I'm also against the instasnap (as opposed to proper bending with horror audio) but disregarding the documented limitations is not the way to go.

[Bremspropeller]

Yeah, right. And there are no over-g inspections carried out ever, because pilots never over-g their jets or stores. It's happenening frequently. Not on a daily basis, but way more often than you'd think.

It's not disregarding limitations. It's about extrapolating exagerated consequences of a one time or two time event that's gonna bend the airframe, but not break it. There's a significant factor between a limitation and "things break". Most structural limitations go into a fatigue life considerations, rather than having direct structural consequences.

In SEA, there were two (IIRC) F-100s throwing their wings during a standard 4g bombing-pullout - mostly because they'd reached their fatigue life earlier than expected. That trigegred a whole IRAN (Inspect And Repair As Necessary) project, as it turned out there were lots of cracks across the fleet. And because people didn't really give a second thought about over-g'ing the airframe in those days. Same with a german 104G - did a gunnery or bombing pass on the Gila Bend range and one wing came off during the pull. Turned out, Lockheed had completely miscalculated the fatigue life of the jet, assuming one 4g pull per sortie. Realistically, ou'd fly about a 4g turn in each corner of the range pattern, though. Note: the wing didn't come off because of an over-g. There was a 104 that survived a double-digit g pullout after a high speed spacial-d event. It didn't fly again afterwards, but it came home in one piece.

Wings did come off jets way more in earlier days - not because of someone accidentally busting a limitation by 1g or two. It happenend because there was no regard for structure-life in the first place and people didn't think much of it. There were a couple of instances of F-8s pulling off their wing. Now the F-8 is nomnally rated for 6gs. But there's an RF-8 that evaded a SAM in SEA and pulled double-digits as well. Probably with some rolling mixed in. It came back to the boat to tell the story.

Mind you that some rollig-g limitations (depending on the jet) are related to stability issues, rather than structure.

There is no horror audio when you over-g the jet. You've got a helmet (possibly earplugs), oxygen flow and a 400kts windnoise which will make you hear no structural noises.

Edited December 8 by Bremspropeller

[Bucic]

Let me rephrase my stance - the wings shouldn't snap as they do. At the same time I routinely fly at 8-9 G with rolling and stores. In my opinion recordings of glaringly premature snapping are needed to have it, freshly, acknowledged by ED for further work. Do I personally feel there's fire on ED's side to get to the bottom of it without waiting for tracks pouring in from users’ side. Probably not.

At the same time let me remind you that it is exactly what led to the wingtip pylon fix - resilient delivery of track recordings with edge conditions.

[Hiob]

1 hour ago, Bremspropeller said:

Yeah, right. And there are no over-g inspections carried out ever, because pilots never over-g their jets or stores. It's happenening frequently. Not on a daily basis, but way more often than you'd think.

It's not disregarding limitations. It's about extrapolating exagerated consequences of a one time or two time event that's gonna bend the airframe, but not break it. There's a significant factor between a limitation and "things break". Most structural limitations go into a fatigue life considerations, rather than having direct structural consequences.

In SEA, there were two (IIRC) F-100s throwing their wings during a standard 4g bombing-pullout - mostly because they'd reached their fatigue life earlier than expected. That trigegred a whole IRAN (Inspect And Repair As Necessary) project, as it turned out there were lots of cracks across the fleet. And because people didn't really give a second thought about over-g'ing the airframe in those days. Same with a german 104G - did a gunnery or bombing pass on the Gila Bend range and one wing came off during the pull. Turned out, Lockheed had completely miscalculated the fatigue life of the jet, assuming one 4g pull per sortie. Realistically, ou'd fly about a 4g turn in each corner of the range pattern, though. Note: the wing didn't come off because of an over-g. There was a 104 that survived a double-digit g pullout after a high speed spacial-d event. It didn't fly again afterwards, but it came home in one piece.

Wings did come off jets way more in earlier days - not because of someone accidentally busting a limitation by 1g or two. It happenend because there was no regard for structure-life in the first place and people didn't think much of it. There were a couple of instances of F-8s pulling off their wing. Now the F-8 is nomnally rated for 6gs. But there's an RF-8 that evaded a SAM in SEA and pulled double-digits as well. Probably with some rolling mixed in. It came back to the boat to tell the story.

Mind you that some rollig-g limitations (depending on the jet) are related to stability issues, rather than structure.

There is no horror audio when you over-g the jet. You've got a helmet (possibly earplugs), oxygen flow and a 400kts windnoise which will make you hear no structural noises.

100% agree

though, as per the title of the thread, I think regardless of the „simulation quality“ of structural issues, being able to pull 20G instantaneously in a Tomcat e.g. (probably worth 40kgs of pull force) without any feedback is part of the problem. I stopped ripping wings of when I started the habit of pulling progressively (like you would/should with the brakes of a motorcycle) instead of yanking like mental. Eventually when I got a FFB stick it became virtually impossible to rip wings (other than intentionally).

[=475FG= Dawger]

2 hours ago, Bremspropeller said:

Yeah, right. And there are no over-g inspections carried out ever, because pilots never over-g their jets or stores. It's happenening frequently. Not on a daily basis, but way more often than you'd think.

It's not disregarding limitations. It's about extrapolating exagerated consequences of a one time or two time event that's gonna bend the airframe, but not break it. There's a significant factor between a limitation and "things break". Most structural limitations go into a fatigue life considerations, rather than having direct structural consequences.

In SEA, there were two (IIRC) F-100s throwing their wings during a standard 4g bombing-pullout - mostly because they'd reached their fatigue life earlier than expected. That trigegred a whole IRAN (Inspect And Repair As Necessary) project, as it turned out there were lots of cracks across the fleet. And because people didn't really give a second thought about over-g'ing the airframe in those days. Same with a german 104G - did a gunnery or bombing pass on the Gila Bend range and one wing came off during the pull. Turned out, Lockheed had completely miscalculated the fatigue life of the jet, assuming one 4g pull per sortie. Realistically, ou'd fly about a 4g turn in each corner of the range pattern, though. Note: the wing didn't come off because of an over-g. There was a 104 that survived a double-digit g pullout after a high speed spacial-d event. It didn't fly again afterwards, but it came home in one piece.

Wings did come off jets way more in earlier days - not because of someone accidentally busting a limitation by 1g or two. It happenend because there was no regard for structure-life in the first place and people didn't think much of it. There were a couple of instances of F-8s pulling off their wing. Now the F-8 is nomnally rated for 6gs. But there's an RF-8 that evaded a SAM in SEA and pulled double-digits as well. Probably with some rolling mixed in. It came back to the boat to tell the story.

Mind you that some rollig-g limitations (depending on the jet) are related to stability issues, rather than structure.

There is no horror audio when you over-g the jet. You've got a helmet (possibly earplugs), oxygen flow and a 400kts windnoise which will make you hear no structural noises.

There is ample evidence that the real F-5 is much, much stronger than the way it is modeled in DCS. I posted this a long time ago.

If you read the linked document carefully you will discover the official fatigue testing program expected 40 9G excursions per 1000 hours (160 excursion per lifetime of 4000 hours) and tested accordingly. The F-5 went through the equivalent of 4 lifetimes of testing or 640 9G excursions with no issues. One can safely assume that expecting 160 9G excursions means at least one excursion per lifetime far in excess if the aircraft/pilot is even able to successfully get to G levels above 9G in the F-5.

I suspect actually generating 13G in a real F-5 isn't physically possible.

There is written evidence that the F-5 is built to withstand severe punishment and it was presented to ED 3 years ago.

The idea that an airplane proven to be able to exceed the published limitations 640 times with no ill effects loses its wings the first time you exceed the limitations is ridiculous on the face of it, yet it remains ignored and even championed by players who paid actual money for a module that was FUBARed intentionally years after its initial release.

We all should have been clamoring for an artificial feel system, and common sense interpretation of the limitations 3 years ago.

Instead, we hear the same " real fighter pilots never exceed the limitations" BS over and over even when presented with clear evidence that it was expected the limits would be exceeded very regularly.

[NineLine]

Hey all, so I think we have improved this somewhat, but one request I have reported was to do proper wing damage vs just snapping the wing. For many years across many sims, the loss of parts was used in lieu of real damage. I think its been used so much, especially in WWII sims people assume its real to have parts break off but in reality, most things short of a large explosion or collision will not do this. 

So currently, and still a 'to do' issue, is the ability to bend the wing instead of break. This means having damage, such as damaged control surfaces, vibrations, etc. 

This is an example of some tests I have been doing with WWII in not allowing the tail to fall off (excuse the videos not lined up that well)

You should still be penalized for abusing the aircraft, and no doubt in my mind its abuse that we are seeing when the wings fall off, but the penalty might be too harsh. 

The other thing we are looking at is adding the artificial feel system in, now if any of you have experience with the Bf 109K-4 in DCS you will see something similar where we simulate control restrictions based on speed, so if you were in a steep dive you would struggle to pull out because the controls become less responsive. I would assume this system for the F-5 would be similar, but adding it might take some time and might be more appropriate for an FFB setup.

So I think it is better, and a little more forgiving, but it still needs some love, until then you meed to be mindful of your speeds and home much force you are using.

Thanks!

[VZ_342]

Mainly I'm looking for if an FFB stick would help reduce either the onset of a snap, or (better) the snap itself. I can't control what ED does or doesn't do, but I can control what I do to reduce the risk of over-stressing the wings.

[Cab]

47 minutes ago, NineLine said:

Hey all, so I think we have improved this somewhat, but one request I have reported was to do proper wing damage vs just snapping the wing. For many years across many sims, the loss of parts was used in lieu of real damage. I think its been used so much, especially in WWII sims people assume its real to have parts break off but in reality, most things short of a large explosion or collision will not do this. 

So currently, and still a 'to do' issue, is the ability to bend the wing instead of break. This means having damage, such as damaged control surfaces, vibrations, etc. 

This is an example of some tests I have been doing with WWII in not allowing the tail to fall off (excuse the videos not lined up that well)

You should still be penalized for abusing the aircraft, and no doubt in my mind its abuse that we are seeing when the wings fall off, but the penalty might be too harsh. 

The other thing we are looking at is adding the artificial feel system in, now if any of you have experience with the Bf 109K-4 in DCS you will see something similar where we simulate control restrictions based on speed, so if you were in a steep dive you would struggle to pull out because the controls become less responsive. I would assume this system for the F-5 would be similar, but adding it might take some time and might be more appropriate for an FFB setup.

So I think it is better, and a little more forgiving, but it still needs some love, until then you meed to be mindful of your speeds and home much force you are using.

Thanks!

I agree it's better but still needs some work. Thanks for letting us know the dev recognize that, too.

[NineLine]

28 minutes ago, VZ_342 said:

Mainly I'm looking for if an FFB stick would help reduce either the onset of a snap, or (better) the snap itself. I can't control what ED does or doesn't do, but I can control what I do to reduce the risk of over-stressing the wings.

I'm still waiting for my FFB stick, so I cant answer that but I am hoping if we add artificial feel in it will, I am also hoping this could be something you could tune yourself with a FFB.

[Raven (Elysian Angel)]

2 hours ago, NineLine said:

You should still be penalized for abusing the aircraft

Oh absolutely, including overspeed damage and stores falling off  

1 hour ago, NineLine said:

I'm still waiting for my FFB stick

Sweet! Which one are you getting? FFB is incredible and you'll wish you got one sooner  

Edited December 9 by Raven (Elysian Angel)
minor grammar fix

[NineLine]

On 12/8/2024 at 7:22 AM, =475FG= Dawger said:

There is ample evidence that the real F-5 is much, much stronger than the way it is modeled in DCS. I posted this a long time ago.

If you read the linked document carefully you will discover the official fatigue testing program expected 40 9G excursions per 1000 hours (160 excursion per lifetime of 4000 hours) and tested accordingly. The F-5 went through the equivalent of 4 lifetimes of testing or 640 9G excursions with no issues. One can safely assume that expecting 160 9G excursions means at least one excursion per lifetime far in excess if the aircraft/pilot is even able to successfully get to G levels above 9G in the F-5.

I suspect actually generating 13G in a real F-5 isn't physically possible.

There is written evidence that the F-5 is built to withstand severe punishment and it was presented to ED 3 years ago.

The idea that an airplane proven to be able to exceed the published limitations 640 times with no ill effects loses its wings the first time you exceed the limitations is ridiculous on the face of it, yet it remains ignored and even championed by players who paid actual money for a module that was FUBARed intentionally years after its initial release.

We all should have been clamoring for an artificial feel system, and common sense interpretation of the limitations 3 years ago.

Instead, we hear the same " real fighter pilots never exceed the limitations" BS over and over even when presented with clear evidence that it was expected the limits would be exceeded very regularly.

Please supply a track of what you are experiencing. Thanks.

[ricktoberfest]

17 hours ago, VZ_342 said:

Mainly I'm looking for if an FFB stick would help reduce either the onset of a snap, or (better) the snap itself. I can't control what ED does or doesn't do, but I can control what I do to reduce the risk of over-stressing the wings.

I just recently switched from a Thrustmaster t16000 (constant wing snaps) to a Logitech g940 force feedback (no wing snaps). Now I don’t fly the F5uch since the wing snaps started but it definitely makes a difference for a couple reasons. 
 

1. the ffb trim options in the special menu means much longer elevator axis so more ability to make smaller movements

2. You can “feel” the movement better

3. it’s harder with ffb to make those sudden movements that can cause wing snap And,

4. Possibly the recent update helped?

[average_pilot]

18 hours ago, NineLine said:

The other thing we are looking at is adding the artificial feel system in, now if any of you have experience with the Bf 109K-4 in DCS you will see something similar where we simulate control restrictions based on speed, so if you were in a steep dive you would struggle to pull out because the controls become less responsive. I would assume this system for the F-5 would be similar, but adding it might take some time and might be more appropriate for an FFB setup.

Let me add that this kind of solutions works perfect with force sensing joysticks since the illusion of needing more effort to move the control surfaces is a real thing, sort of like a FFB in that regard.

IMHO those are the only two kind of joysticks worth having for serious simulation. Each type in its own flavor allows for an interaction with the software that's totally absent in a mechanical joystick that you can just move to the extremes using always the same force no matter what is happening in the sim.

[RustBelt]

19 hours ago, VZ_342 said:

Mainly I'm looking for if an FFB stick would help reduce either the onset of a snap, or (better) the snap itself. I can't control what ED does or doesn't do, but I can control what I do to reduce the risk of over-stressing the wings.

I have always run a Sidewinder FFB2 and have never snapped a wing. 

[=475FG= Dawger]

3 hours ago, NineLine said:

Please supply a track of what you are experiencing. Thanks.

No thanks.
 

I have already wasted many hours researching the aforementioned post and several others as well as posting multiple tracks and videos of the behaviors of concern only to be told to stop doing things that make the wings break. 
 

I don’t think it was you but honestly don’t remember. 
 

The F-5 desperately needs artificial control feel modeled for the pitch axis and it needs someone who can correctly interpret and apply the roll limitations. If the change notes someday include those items, I would be happy to test them. 
 

Its good see acknowledgement of catastrophic failure as unrealistic and hopefully that will be corrected in the short term. 

[Bucic]

4 hours ago, NineLine said:

Please supply a track of what you are experiencing. Thanks.

I could do that but first please consult the devs on the agreed loadout, OK? I may produce some track but to spend more time on it I'd need a starting point.

[Bucic]

Did 2 trials with

80% fuel

2 x AIM-9P

center tank 150

IAS 450-500

Xbox controller RAW pitch axis (no curvature, no saturation) - so I'm guaranteed to give a faster input than any joystick rider.

Roll axis still with some 15% curvature and 88% saturation. I wanted to isolate the roll a bit at first.

 

Results:

Some 10 attempts with sharp full pitch input coupled with roll, sometimes with yaw, with no wing snaps. You can witness multiple 9+G onsets attained in 0.5s or maybe even faster.

What I did avoid conciously was preceeding the positive incursions with negative ones. Also, at one point after several "unsuccessful" attempts I immediately got a wing snap... when I use an rudder input opposite to the turn.

You can also see I attempted abrupt pitch commands having it set up with preceeding rudder kicks.

attempt 01

 

 

attempt 02

 

Both tracks can be found here

https://1drv.ms/f/c/66cf2646234394fc/Eq2pxziu5-9MrFLEnhqFCHUB9arWL83TFma8zE0adhUEjw?e=WqYEVj

Additionally bug F-5E REM notbugtest elevator deflection rate mach 1.trk shows that the module has some limitations on surface deflection under aerodynamic load. Which is good. Puzzling to me is the clear discrepancy between up and down pitch. Pitch down (stick pushed) makes the surface deflect faster.

Edited December 9 by Bucic

[RustBelt]

Could be designed that way for stall recovery? 

[NineLine]

5 hours ago, =475FG= Dawger said:

No thanks.
 

I have already wasted many hours researching the aforementioned post and several others as well as posting multiple tracks and videos of the behaviors of concern only to be told to stop doing things that make the wings break. 
 

I don’t think it was you but honestly don’t remember. 
 

The F-5 desperately needs artificial control feel modeled for the pitch axis and it needs someone who can correctly interpret and apply the roll limitations. If the change notes someday include those items, I would be happy to test them. 
 

Its good see acknowledgement of catastrophic failure as unrealistic and hopefully that will be corrected in the short term. 

Well, what I wanted to know is if you have tried it with the changes, as well as what loadouts you were using, are these tests in real conditions or are you just jumping in, getting a bunch of speed and pulling back as hard as you can? Some damage has been introduced by way of hydraulic failures. Anyways, fair enough if you are not interested in checking, that is fine.

3 hours ago, Bucic said:

Did 2 trials with

80% fuel

2 x AIM-9P

center tank 150

IAS 450-500

Xbox controller RAW pitch axis (no curvature, no saturation) - so I'm guaranteed to give a faster input than any joystick rider.

Roll axis still with some 15% curvature and 88% saturation. I wanted to isolate the roll a bit at first.

 

Results:

Some 10 attempts with sharp full pitch input coupled with roll, sometimes with yaw, with no wing snaps. You can witness multiple 9+G onsets attained in 0.5s or maybe even faster.

What I did avoid conciously was preceeding the positive incursions with negative ones. Also, at one point after several "unsuccessful" attempts I immediately got a wing snap... when I use an rudder input opposite to the turn.

You can also see I attempted abrupt pitch commands having it set up with preceeding rudder kicks.

attempt 01

 

 

attempt 02

 

Both tracks can be found here

https://1drv.ms/f/c/66cf2646234394fc/Eq2pxziu5-9MrFLEnhqFCHUB9arWL83TFma8zE0adhUEjw?e=WqYEVj

Additionally bug F-5E REM notbugtest elevator deflection rate mach 1.trk shows that the module has some limitations on surface deflection under aerodynamic load. Which is good. Puzzling to me is the clear discrepancy between up and down pitch. Pitch down (stick pushed) makes the surface deflect faster.

Thanks for this, I would be interested in real conditions, during a dog fight, etc. I remember the original issue and it was super sensitive to breakage. So tracks from fights or other situations like that.

[Bucic]

7 hours ago, NineLine said:

Well, what I wanted to know is if you have tried it with the changes, as well as what loadouts you were using, are these tests in real conditions or are you just jumping in, getting a bunch of speed and pulling back as hard as you can? Some damage has been introduced by way of hydraulic failures. Anyways, fair enough if you are not interested in checking, that is fine.

Thanks for this, I would be interested in real conditions, during a dog fight, etc. I remember the original issue and it was super sensitive to breakage. So tracks from fights or other situations like that.

Oof, that's were we'd need 

https://m.youtube.com/@Bullet4MyEnemy/videos

I'll ask for some info.

Seeing how much overlimit conditions it took to break the wings during my flights the wings would never break with my flying.

On my end I'm going to expand into:

1. Abrupt pitch up preceded with slight abrupt pitch downs

2. RAW roll input with no curvature and no saturation.

BTW, mark my words, deadzone settings people are using are going to be a factor in this bug

[Bucic]

OK, I managed to snap the wings at 5G and... 1.5 deg AoA (TacView). The snap was preceeded by 2 high G pitch commands. Track, video as well as a TacView recording coming along. It turns out I bought TacView Advanced on sale 9 years ago

Is there a way to do a custom plot of G vs AoA in TacView?

On limitations of TacView

https://tacview.fandom.com/wiki/Formulas#G-forces

You can bookmark this OneDrive link, today's session uploaded:

https://1drv.ms/f/c/66cf2646234394fc/Eq2pxziu5-9MrFLEnhqFCHUB9arWL83TFma8zE0adhUEjw?e=WqYEVj

 

 

 

 

Edited December 10 by Bucic