can not reproduce Severe Flight Modeling Glitch. Post Stall behaviour.

[twistking]

I managed to stall the viper in a Dogfight and went full Afterburner to save her. I then got the plane in a hovering state, with thrust cancelling out gravity. However the plane was relatively unresponsive to inputs while it was slowly rocking from side to side, while still maintaining altitude like a helicopter. Full stick forward did nothing, rudder had effect, but could not bring her out of her limbo. The akward thing was, that while i was hovering in position, my thrust vector was not downwards but slightly off, or in other words my nose was not pointing directly upwards. Yet i was still hovering in position and gently rocking from left to right as if grabbed by some pre-baked stall-animation.

Only when the fuel was nearly empty, did i somehow manage to exit this hover-stall-state. I don't know how exactly.

 

I've saved the track file, however on my pc it did not play correctly. Maybe because i used some time acceleration during replay (it happened +20min in the mission)?!

Also it's too big to upload here, even when zipped (7mb).

 

If this problem with weird post stall afterburner glitch hovering is not yet documented, i can maybe upload the track elsewhere and provide a link.

 

[Deano87]

At what altitude did you stall? And are you saying it didn't lose any altitude while hanging there?

The F-16 should in extreme cases get stuck in a nose high "falling leaf" deep stall which requires use of the MPO (Manual Pitch Override) switch to be able to get out of. If you were nose high, but you were actaully losing altitude, then this is what you were in, if you were literally hovering in the sky, then no idea.

Edited January 4, 2021 by Deano87

[twistking]

31 minutes ago, Deano87 said:

At what altitude did you stall? And are you saying it didn't lose any altitude while hanging there?

The F-16 should in extreme cases get stuck in a nose high "falling leaf" deep stall which requires use of the MPO (Manual Pitch Override) switch to be able to get out of. If you were nose high, but you were actaully losing altitude, then this is what you were in, if you were literally hovering in the sky, then no idea.

 

 

i think i lost quite a lot of altitude in the beginning, but when i got into the hover i was only few thousand feets high. i would guess 3000, but did not check. and yes, at that moment i was completely stable, not loosing altitude. only rocking a bit from side to side and maybe slowly moving a sideways.
i did not know about the f-16 deep stall you mention. it seems to me, that i was indeed in such a state, only that after a time my altitude stabilized and i was in that weird hover. the aircraft was very light at that point: no ordnance left, no gun ammo left, no ext tanks, no tgp and full afterburner does empty the internal tanks quite efficiently.

 

maybe this stall characteristic is indeed hardcoded (lut-based) and the programmer did not calculate correctly for a plane so light, so that the aircraft wanted to gain altitude (by thrust alone) but could not, because it was "captured" in that hard-coded stall behaviour.

when i was nearly completely empty on fuel i somehow managed to "roll out" of the stall (i think by pitching with the rocking movement), the aircraft tipped over and i managed to get it back to stable flight only some meters above the water surface. the "rolling out" also felt natural to me. it went from that weird nose up stall to a more typical stall behaviour which was easy to safe with the high thrust that aircraft produces.

 

so the big "glitch" was maybe only me hovering with constant altitude (for quite a long time by the way - an enemy f-5 had time to do several gun passes on me, but could not hit).

the weird nose up stall and the final roll out would seem plausible to me now, if it wouldn't have been for the minute of hovering with 80° nose up attitude in between...

 

[Deano87]

I don't think the deep stall behavoir is hardcoded, every time I've got into it, its been very dynamic and unpredictable and different. I just think perhapse you managed to settle into some weird balance point. Have you treid re-creating it?

[twistking]

28 minutes ago, Deano87 said:

I don't think the deep stall behavoir is hardcoded, every time I've got into it, its been very dynamic and unpredictable and different. I just think perhapse you managed to settle into some weird balance point. Have you treid re-creating it?

 

i was hovering for so long that i could check the outside camera and it was clear to me, that this was not physically correct. for a balance thrust would need to go roughly downwards but i was hanging there slanted. maybe 80° nose up, maybe 70°. also i did not significantly alter altitude while burning through most of my fuel. i played enough KSP to know for sure that that was not rocket science.

 

i can't recreate it unfortunately, because i had the viper from the free trial and on steam the trial ended yesterday. if someone else wants to try to recreate: i had expanded all my mubnitions, missiles and gun and was at roughly 50% percent internal fuel.

i entered the stall on full military power and went full AB when realizing the severity of the stall (high nose stall, like deano87 mentioned in his reply). i did only minor control inputs because i was assuming the f-16s high thrust/weight ratio and fcs would save me anyway. the afterburner then slowed my descend to a stop which left me in this weird slanted hover.

[Deano87]

I'll have a go... obviously you had everything like game flight model etc disabled in the options?

[twistking]

5 minutes ago, Deano87 said:

I'll have a go... obviously you had everything like game flight model etc disabled in the options?

 

yes, absolutely.

[twistking]

@moderators: i see, i got the "cannot reproduce" batch. would it help, if i provide a link to my trackfile? you'd have to sit through 20min of me just cruising around testing radar modes and aim-9 HOBs before the stall happens. i cannot open the trackfile myself (i guess) because i have no access to the module any more. otherwise i could have tried to capture a video.

[BIGNEWY]

Yes if you have a track, although long tracks may not playback correctly it is worth a try. Please link and we will take a look.

 

If you are able to reproduce on a airstart mission it would be best

 

thanks

[Center_Mass]

On 1/11/2021 at 6:54 AM, BIGNEWY said:

Yes if you have a track, although long tracks may not playback correctly it is worth a try. Please link and we will take a look.

 

If you are bale to reproduce on a airstart mission it would be best

 

thanks

 

I have a trk file but it's too large for the forum ~ 7MB,  should I host it on my Google drive? I also attached the Tacview file of it.

 

Tacview-20210602-170356-DCS-21 Arena for BFM ACM Training 1vs1 2vs2 4vs4 XX vs XX 30°C 10h30 restart 4h 21.04 c1.zip.acmi

Edited June 2, 2021 by Center_Mass

[BIGNEWY]

we need a track replay if possible, if the bug is there it should be repeatable on a mission with just the aircraft in. 

 

Please note the flight model is W.I.P and a review is already underway. 

 

thanks 

[Center_Mass]

Just now, BIGNEWY said:

we need a track replay if possible, if the bug is there it should be repeatable on a mission with just the aircraft in. 

 

Please note the flight model is W.I.P and a review is already underway. 

 

thanks 

 

F-16_post_stall_hover_trk. Just want to be sure you are aware of the issue.  I don't think I've seen this behavior on a jet without damage.

 

Looking forward to the new updates and fixes.

 

Regards,

CenterMass

 

 

[uri_ba]

Deep stall on the viper is pretty nasty situation, and you cannot exit it without following the checklist.
if you "catch it in time" (i.e before it fully develop, you can still get out of it).

basics of a deep stall is that the plane settles into an AOA such that the elevators are in "dead air" in the wake for the wing flow separation. it is more common in T-Tail aircraft.
however, the Viper is unique in the way it's FLCS tries to correct. so it will start rocking back and forth but will never come back to normal flight.

you should hold the MPO and "help" the oscillation to develop (i.e pull back when the nose comes up, and push when nose comes down).  eventually, nose will plunge down for the elevator to be effective again and plane will be in normal controlled flight.

MPO however is a double edged sword. it overrides the FLCS pitch authority completely. and will be misused in MP dogfights if implemented correctly (hadn't tried it in DCS yet)

[Center_Mass]

3 hours ago, uri_ba said:

Deep stall on the viper is pretty nasty situation, and you cannot exit it without following the checklist.
if you "catch it in time" (i.e before it fully develop, you can still get out of it).

basics of a deep stall is that the plane settles into an AOA such that the elevators are in "dead air" in the wake for the wing flow separation. it is more common in T-Tail aircraft.
however, the Viper is unique in the way it's FLCS tries to correct. so it will start rocking back and forth but will never come back to normal flight.

you should hold the MPO and "help" the oscillation to develop (i.e pull back when the nose comes up, and push when nose comes down).  eventually, nose will plunge down for the elevator to be effective again and plane will be in normal controlled flight.

MPO however is a double edged sword. it overrides the FLCS pitch authority completely. and will be misused in MP dogfights if implemented correctly (hadn't tried it in DCS yet)

 

Did you watch the Tacview file the issue starts at 15:50? This an issue of a jet being damaged and not losing hardly any altitude due to a nose high attitude that is sustained. This may be accurate since the plane is damaged, but would just like to have it evaluated by the Dev's / SME's.

 

Players trying to use MPO in a DF wouldn't get much help from it since they would likely depart the jet from controlled flight quite often. The only place I could see MPO useful is during slow flight for more AoA which would exacerbate the issue.

 

Also this article "Recovering From Deep Stalls And Departures" cited in the F-16 Wiki article has a lot of information for what a deep stall is, and how to get out of it.

 

You can see from the Tacview that it appears that the current DCS F-16 too much nose up authority in the deep stall.

[LJQCN101]

You basically won't be able to use the MPO in a dogfight unless you use it in the negative G direction. It can only override AOA limiter (actually sets AOA feedback to 0) when AOA > 35 degrees.

 

Here's what MPO OVRD did to the longitudinal FLCS:

EDIT: The current MPO implementation seems to be correct, except that the maximum allowed negative G should be -3G, instead of the current -2G.

Edited June 7, 2021 by LJQCN101

[Center_Mass]

9 hours ago, LJQCN101 said:

You basically won't be able to use the MPO in a dogfight unless you use it in the negative G direction. It can only override AOA limiter (actually sets AOA feedback to 0) when AOA > 35 degrees.

 

Here's what MPO OVRD did to the longitudinal FLCS:

EDIT: The current MPO implementation seems to be correct, except that the maximum allowed negative G should be -3G, instead of the current -2G.

 

 

Rumor has it that MPO is not implemented yet, which would make sense since the FM is still in development. I don't know whether it is or isn't though.

 

I would also expect user to exploit systems, like MPO, in dogfights or other regimes of flight. Players use overrides designed for safety to get performance gains where they would probably damage the jet or cause it to depart. I hope they implement, or update, how the MPO works to what you show if what you show is accurate.

[Dee-Jay]

MPO do only allow elevons full down authority. No cheat in dough fight possible (unless intention is to depart in a negative AoA loss of control ;)).

[BIGNEWY]

Please if possible

 

I need a short track replay example in single player that clearly shows a problem here. 

 

The track should be as short as possible to show the problem. 

 

thanks

[Center_Mass]

5 hours ago, BIGNEWY said:

Please if possible

 

I need a short track replay example in single player that clearly shows a problem here. 

 

The track should be as short as possible to show the problem. 

 

thanks

 

I have not seen this on a jet that was not damaged. Don't think I can reproduce it single player. But the issue happens at 13:35 track time and is on the Ares F-16.

 

F-16_post_stall_hover_trk

[BIGNEWY]

5 minutes ago, Center_Mass said:

 

I have not seen this on a jet that was not damaged. Don't think I can reproduce it single player. But the issue happens at 13:35 track time and is on the Ares F-16.

 

F-16_post_stall_hover_trk

 

thanks for trying, but the track replay does not seem to show the viper, just a gun emplacement. 

[Sleepy]

27 minutes ago, BIGNEWY said:

 

thanks for trying, but the track replay does not seem to show the viper, just a gun emplacement. 

The track seems to work fine for me, this happens at about the 13:39 mark

[Sleepy]

Here are two shorter tracks:

 

The first is a viper with all pylons, and it hit a higher overall AoA but couldn't sustain the 60*

 

The second is a clean viper that hit a lower max AoA, but sustained 60*

 

 

OverAoA.trk OverAoA2.trk