Jump to content

Pitching moment near 90 deg. AoA


Maverick Su-35S

Recommended Posts

Hello everyone,

 

 

I know that some don't like to see me very often while others share my thoughts and knowledge and know what I'm talking about.

 

The MIG-21 is known to be an aircraft with quite a good longitudinal static stability margin (great arm between CG and general center of lift or pressure (CP)). Now this is perfectly reproduced by our sim up to a given AoA, after which the things start to go wrong.

 

The longitudinal static stability (or otherwise referred to as pitch stability), as long as it's still positive, not neutral or negative (no known fighter aircraft has it negative even if you hear some ignorant saying: "Unstable aircraft, Unstable aircraft!", don't bother with him, cause he doesn't know what he's saying...), should allow an aircraft to keep reducing it's angle of attack as close to 0 lift as possible, when the elevator is at 0 (neutral).

 

All modern fighter aircraft are being designed to have a reduced longitudinal static stability margin or very close to 0 (relaxed static stability aircraft) in order to obtain much better flight characteristics by having the elevator take part in lifting the aircraft, so a more reduced longitudinal static stability aircraft would have a more downwards deflected elevator which produces lift in the same direction as the wing, thus increasing the total lift.

 

So far, there isn't any aircraft that would be unstable, because after all, unstable IS unstable and no computer can bring it back once it passed over a certain limit and might mostly be able to hold it only through oscillations (it's like trying to keep a pen in equilibrium on your finger) and no such thing exists for flyable aircraft, but anyway this is not our issue here and the MIG-21 doesn't do this either, I just wanted to point out the 3 different situations so no one could confuse them.

 

The problem is that after some patches ago (don't know how many), the MIG-21's FM had been tweaked and some things have gotten better, but other things are not doing the same, and here's how it is:

 

1. The pitching moment has some strange behavior between certain AoA values. The first one would be exactly where the wings start stalling, which is 32-33 deg. AoA on the AoA indicator or 20 deg. of real AoA as measured more correctly. Starting from the stalling AoA and continuing to slightly increase it, an abrupt and strong pitch stability reduction occurs and according to the elevator deflection that the aircraft has at that point, the AoA accelerates rapidly to about 30-40 deg. or so where it abruptly stops!

 

This could only be imagined as the CP would sharply move quite forward and very close to the CG (not beyond which would mean "unstable") in the range of one or 2 degrees of AoA and then coming back almost close to it's original position when the AoA reaches about 30-40 as I said.

 

Now, although there might be some heavy flow detachments starting from the trailing edge towards the leading edge (on the upper surface for positive AoA), while part of the leading edge might still retain flow attachment due to the vortexes generated between the leading edge and root which would translate into the high loss of longitudinal static stability margin because the CP moves forward towards the vortexes, the fact that when reaching about 30-35 deg. AoA the stability margin becomes abruptly very high, which isn't right, even if we'd talk about the vortex breakdown that might occur and which would increase the pitch down moment, still the transition shouldn't be so rough. So, all I'm saying is that the transition or traveling of the center of lift or pressure "CP" shouldn't be so abrupt and should be as smooth as it can be found for the F-15 which should have quite similar static stability characteristics in reality.

 

Another and much greater issue is that when reaching 90 deg. of AoA, which can only be reached through stall spins (which is absolutely normal for such a highly stable aircraft), after reducing the yaw rate to 0 the aircraft trims itself to 90 deg. of AoA like if it's statically relaxed there and won't budge in pitch whatever you'd try. This is also abnormal..., normally the higher the AoA (after the stall occurs) the higher the stability margin and therefore the higher the stabilizing moment should be towards reducing the angle of attack, but it seems that at 90 deg. there is no more stability margin left at all.

 

2. Indeed the stalled lift is no longer 0 cause that wasn't correct, but it's still very small compared to what it should've been, at least from my personal point of view. As you increase AoA and go passing through critical and start stalling, from the maximum achieved G-load, it drops to about 20-25% of what it had been shortly before stall, which still looks not OK. There should be some more...! There are a lot of CL vs AoA diagrams to be found for whole 3D wings, not just for airfoils, that could give a general feeling of how much should the lift drop after the stall and with what slope it should continue to rise as the AoA continues to rise throughout and beyond the stall up to 40-45 deg. from where it naturally starts dropping towards 0 as the AoA reaches 90. Furthermore, for a delta wing and mostly for highly swept deltas, the stall should be more docile/gradual due to the strong vortexes that are being generated between the leading edge's root and the fusealge and also, before the vortex breaks up (some 5-10 more deg. of AoA (depends from a wing to another) since the stall occurred) the lift drop beyond stall should be smaller than for a straight, high aspect ratio wing...! I tell these facts because this is my domain and there is proof for this. I'm not complaining about this, but..., there should be a lot more lift remaining after the stall occurs.

 

Please test it:

MIG-21 80..90 deg. AoA gives 0 pitching moment.trk

 

Thank you!


Edited by Maverick Su-35S

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

.

 

So far, there isn't any aircraft that would be unstable, because after all, unstable IS unstable and no computer can bring it back once it passed over a certain limit and might mostly be able to hold it only through oscillations (it's like trying to keep a pen in equilibrium on your finger) and no such thing exists for flyable aircraft, but anyway this is not our issue here and the MIG-21 doesn't do this either, I just wanted to point out the 3 different situations so no one could confuse them.

 

Sorry, I haven't yet fully read your analysis on the mig21. But I must point out that this is incorrect.

Trying to keep a pen in equilibrium on its tip is perfectly doable ! There are tons of robots/machines and human capable of that ! The pen is like you said unstable and will fall down if left on its own. But if given the right inputs of force at its base it can be kept upright!

This is a case of an unstable system that as been stabilised by introduction of an active control ( usually in the form of a pid controller). Look up control theory.

 

It is the same for aircrafts. Some advanced fighters might be unstable (static margin slightly negative) or marginally stable (static margin positive but very close to 0),but the addition of a complex flyby wire system between the pilot and the control surfaces can stabilise them! Of course they will be very hard, if not impossible to fly without the flight controller .(press S in the su27 to see what I mean )


Edited by luckyhendrix
Link to comment
Share on other sites

Sorry, I haven't yet fully read your analysis on the mig21. But I must point out that this is incorrect.

Trying to keep a pen in equilibrium on its tip is perfectly doable ! There are tons of robots/machines and human capable of that ! The pen is like you said unstable and will fall down if left on its own. But if given the right inputs of force at its base it can be kept upright!

This is a case of an unstable system that as been stabilised by introduction of an active control ( usually in the form of a pid controller). Look up control theory.

 

Look, you don't have to teach me about flight dynamics or stable/unstable systems cause I know them better than you think! You are wrong and contradict yourself when you talk about an unstable system which behaves like a stable one..., that means you didn't understand what instability is. By saying that you or a machine can keep a pen "upright" with constant forces, without oscillations (and I've already mentioned this), then it means you don't know what you're saying. I say again: When it passes over a certain limit, not even the robot/computer or whatever can turn it relatively stable again. You try to prove that you know things better, when actually and honestly, you don't...!

 

This is one example that I know about which has 2 degrees of freedom (I hope you know what a degree of freedom is), and a programmed arm will keep it in a RELATIVE upright position, but not absolutely upright as you told/think!. Watch from 0:55:

 

 

Now about fighter aircraft...! Do you know when a fighter aircraft actually becomes unstable (when the static stability margin becomes negative)? How would you describe the control surfaces deflections that control the stability of such an aircraft?

 

Indeed the static stability margin may become negative, but only in some conditions that you definitely didn't mention which leaves me with the thinking that you actually don't know about. You are talking to me about fly-by-wire and active control systems and about control theory and such..., that are being used to make an unstable system behave relatively like a stable one, but it more seems that you can't handle such things, or at least this is what you make me think!

 

It is the same for aircrafts. Some advanced fighters might be unstable (static margin slightly negative) or marginally stable (static margin positive but very close to 0),but the addition of a complex flyby wire system between the pilot and the control surfaces can stabilise them! Of course they will be very hard, if not impossible to fly without the flight controller .(press S in the su27 to see what I mean )

 

Speaking of witch..., to give you a hint that your examples are contradictory to what you say, I was able to fly the Su-27 with ASC direct control ON, do well controlled manoeuvres including shooting other aircraft with the gun in a dogfight, and land as smoothly as normal. Now you should also try and fly and control the Su-27 in pitch with the ASC direct pitch control on and tell if that's unstable while flying..., if so, where..., if not, why? If you can't answer this question correctly, then we have nothing more left to talk about on this subject!

 

Don't get me too wrong, but I actually started to loose my patience with those trying to look smarter before they know what they're actually saying, sorry...!

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

You are playing on words, I didn't mean that the inverted pendulum can be kept perfectly upright at all times.But the system has been stabilized has shown by your very own video...

 

Concerning the su-27 I didn't say it is an unstable aircraft. But it must have a tiny static margin, with the ASC on, it is perfectly stable. With the ASC off you can easily give controls input that send it in an unrecoverable situation.

 

And cut the attitude, I'm not going to bother read the rest of your post like I was going to. Because you're going to be a smartass about it as usual ...

Link to comment
Share on other sites

  • ED Team
You are playing on words, I didn't mean that the inverted pendulum can be kept perfectly upright at all times.But the system has been stabilized has shown by your very own video...

 

Concerning the su-27 I didn't say it is an unstable aircraft. But it must have a tiny static margin, with the ASC on, it is perfectly stable. With the ASC off you can easily give controls input that send it in an unrecoverable situation.

 

And cut the attitude, I'm not going to bother read the rest of your post like I was going to. Because you're going to be a smartass about it as usual ...

 

You are not right. Su-27 is slightly unstable in a wide range of GW. The subsequent Su-27 modifications have even more instability. The point is that the more unstable is the design the more lift (and not negative lift as for the stable plane) can be provided to balance the plane at high CL.

 

Anyway, my notes do not concern to your observation on MiG-21 stability :).

Ніщо так сильно не ранить мозок, як уламки скла від розбитих рожевих окулярів

There is nothing so hurtful for the brain as splinters of broken rose-coloured spectacles.

Ничто так сильно не ранит мозг, как осколки стекла от разбитых розовых очков (С) Me

Link to comment
Share on other sites

You are not right. Su-27 is slightly unstable in a wide range of GW. The subsequent Su-27 modifications have even more instability. The point is that the more unstable is the design the more lift (and not negative lift as for the stable plane) can be provided to balance the plane at high CL.

 

Anyway, my notes do not concern to your observation on MiG-21 stability :).

I didnt know what is exactly the static margin of the the su27. I guessed it had to be tiny or negative by the way it handles. But yes if you say the su27 is unstable it seems logical.

 

And thus we are saying the same thing: "aircrafts naturally unstable exists" .

Link to comment
Share on other sites

Modern airplanes are naturally unstable, F-14 and mig-29 onwards, they use their flybywire systems to keep them in the air, that makes them a lot more manoeuvrable, since they more redily lose lift.

 

the mig-21 on the other hand, is extremely stable, think of it more like an arrow or a dart than an airplane. Because the entire plane is a lifting body, not just the wings, and it's long and narrow, it naturally flies in a straight line, so it wants to return from any amount of AOA just by itself.


Edited by Hadwell

My youtube channel Remember: the fun is in the fight, not the kill, so say NO! to the AIM-120.

System specs:ROG Maximus XI Hero, Intel I9 9900K, 32GB 3200MHz ram, EVGA 1080ti FTW3, Samsung 970 EVO 1TB NVME, 27" Samsung SA350 1080p, 27" BenQ GW2765HT 1440p, ASUS ROG PG278Q 1440p G-SYNC

Controls: Saitekt rudder pedals,Virpil MongoosT50 throttle, warBRD base, CM2 stick, TrackIR 5+pro clip, WMR VR headset.

[sIGPIC][/sIGPIC]

Link to comment
Share on other sites

  • 1 month later...

Look guys, I apologize for my bad mood, but I just don't know how to explain (at least from what I learned, not something out of my own beliefs) that there are 3 types of system stability (in 2 separate categories), which have been defined as a convention which everyone agrees with.

 

The two categories are: static stability and dynamic stability! Each of them (separate or together) can provide one of the three stability regimes: stable, neutral stability (known as relaxed stability) and unstable. So there's a total of 6 possible situations:

 

a) statically stable, statically neutral/relaxed, statically unstable;

b) dynamically stable, dynamically neutral/relaxed and dynamically unstable.

 

The dynamic stability is defined through a series of conventional (known) modes of oscillation, and is almost always present next to the static stability of a system.

 

Our concern only regards the static stability of an aircraft so we should not even think of the dynamic stability (which is more related to time and frequency) which indeed might superimpose with the static stability at some random situation or make a confusion between the two!

 

As far as I know, an aircraft is told to be statically stable if it has the tendency towards decreasing the AoA, WHEN the pitch control (horizontal tail or canard) is at it's null position (a position of 0 deflection told by the manual) or the AoA would go to another equilibrium value according to a new elevator position. Hence, this is what we call a statically stable aircraft (longitudinal stability as our concern), or an aircraft which still has a tiny fraction of static stability still available.

 

If the elevator/canard is at 0 deflection and the AoA remains at rest in any preset position, then the aircraft is told to be statically relaxed (neutral static stability), not having the tendency to vary any new AoA value when the elevator is brought to 0.

 

The third situation, which everyone seems to share so rapidly/easily is about having the elevator at 0 deflection and the aircraft's AoA tends to increase by itself uncommanded towards the critical AoA and beyond..., is when the aircraft is told to be statically unstable and won't return to a controlled flight whatsoever..., and THIS IS NOT the case with the Flanker nor the F-16 eighter (even if you don't believe it at first). Someone here said that even the F-14 and MIG-29 are unstable aircraft too, so..., this leaves me out of any other word!

 

Please remember this: "If the AoA will always find another equilibrium value according to a new elevator position and will return to the original value when the elevator is brought to it's original position, THAT AIRCRAFT IS STILL STATICALLY STABLE", it is not unstable, which is the case with our Flanker too and it's very easy to test this in game without telling a single word about it. It is still within a statically stable flight, what the heck...!

 

If the AoA won't stop at a random equilibrium value when the elevator is brought to 0 deflection, which is the case for statically relaxed condition, and keeps on building up continuously even with the consequence of going past beyond 90 deg. AoA, that's the only condition for an aircraft to be called unstable, otherwise the word "unstable" is used by many as a pretext for the high pitch response of the aircraft for slight elevator deflections (which seems very unusual to them) and so they to call it unstable for that cause, while in reality it is still flying with some static stability left because the AoA stops at some point according to a new elevator position and the aircraft is fairly controllable and still flying more or less well (according to the new AoA due to elevator deflection), otherwise the AoA would continue building up until the plane would be flying tail forward. I don't know how to better explain this, but it is just as simple as it is defined.

 

I don't know how would someone else (who has the knowledge) describe aircraft static instability if anything else than something that won't find an equilibrium on it's own unless there is a CONSTANT intervention (from an automated system or man) that would always and constantly try to "find" an elevator position for which the AoA would stay in a so called "equilibrium", yet that would require the elevator to constantly make quite noticeable adjustments in order to not let the AoA vary (as it tends to do) and in that kind of situation you'll ALWAYS see that aircraft flying in a pitch oscillatory trajectory, and no matter how small the amplitude might be, it will be exactly as in the case with the pendulum, the computer just CAN'T find an elevator position (cause there isn't any) for which the AoA wouldn't vary anymore..., the same way, a robot CAN'T find a constant position for which the pendulum would find equilibrium (unless there's friction of course). By definition, THAT'S INSTABILITY FOLKS...! Nothing of the latter is attributed to the Flanker's behavior with ASC ON, neither in real flight or in DCS.

 

The only purpose of the flight control system that filters the pitch rates, AoA rates and AoA values is to give the pilot a lot less headache while flying the aircraft, because it's a fighter pilot in a fighter aircraft and wants to be more concerned on how to defeat the enemy and fly the plane to the edge of it's flight envelope carefree rather than fighting more with his own plane in order to have coordinated and well controlled manoeuvres even when the aircraft is still statically stable..., THAT'S the reason of the fly by wire electro-mechanical system in general, to help the pilot have more precise control, NOT because the plane would be unstable! Take for example the CAS on the F-15 or the SAS on the A-10, try them off and see the difference..., and they are all statically stable designs! It will lead you to the feeling that it's "unstable" in pitch or yaw with those stability systems off, but the planes would have the same static margin that they were born with...!

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

TL;DR because of current situation ... Are you saying these plants won't fly a proper phugoid in the sim when trimmed?

 

Thread went off-topic fast. He was complaining about the MiG-21's tendencies in the stall- if you stall it out the MiG rapidly pitches up to ~30 degrees AoA and 'sticks' there, requiring nearly full down elevator to recover. This seems strange, as it would imply that the centre of pressure moves forward when the stall occurs, something that seems at odds with the aerodynamic configuration of the MiG.

 

 

On the topic of stability, I would submit that it is possible to control a statically unstable aircraft so long as its AoA does not go beyond the point where the controls cannot overcome the vehicle's tendency to increase AoA further. This is certainly the case with space rockets, which essentially always have their aerodynamic centre of pressure well ahead of their centre of mass and manage to fly just fine...

.
Link to comment
Share on other sites

Thread went off-topic fast. He was complaining about the MiG-21's tendencies in the stall- if you stall it out the MiG rapidly pitches up to ~30 degrees AoA and 'sticks' there, requiring nearly full down elevator to recover. This seems strange, as it would imply that the centre of pressure moves forward when the stall occurs, something that seems at odds with the aerodynamic configuration of the MiG.

....

 

Yes, I started the topic in order to make clear the problem that the MIG-21 seems to have regarding the pitch accelerations (in aerodynamics/flight dynamics, this is directly reflected into Cm (pitching moment coef.)) variations between certain AoA. If needed, I'll provide a chart with the pitching moment (or Cm if wanted) variation with alpha on the MIG-21 for full aft stick (with an initial pitch trim position) in order to illustrate the way it looks "from outside".

 

Now I don't want to look wise, but I really felt that I can't even begin to talk without telling what I know (and I'm not some dumb who just showed up sharing his basic knowledge about flight mechanics and mechanics in general) about how a statically stable aircraft (the MIG-21 is a strong example) would react in every condition.

 

Thanks!

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

  • 2 weeks later...
Thank you for your report.

 

Aircraft behavior near critical and on supercritical AoA will be improved and available in the next patch.

 

THANK YOU "Dolphin887"! It's so good to hear that!

 

The pitching moment or pitching rate accelerations as behavior of the MIG-21 when passing through stall AoA (at least for positive stall AoA) with full negative elevator deflection (full aft stick) and pitching moment at or near 90 deg. AoA must be re-evaluated, because the tracks that I've provided in the first post already proves it!

 

 

Best wishes! Keep up the good work!

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

  • 8 months later...

Hi again,

 

Good progress guys, now the 90 deg.AoA stuck behavior has been eliminated and the plane responds like a real one and in the same time the strange and very abrupt pitching moment variations have also been eliminated between the stall AoA and some 10..15 deg. more, thus the pitching moment variation for a constantly held elevator deflection as the aircraft passes through the stall AoA towards the maximum AoA achievable is more gradual and looks quite realistic and also the plane develops some outstanding pitch-roll oscillations in certain conditions during which is unprecedentedly realistic, but still, some other problems emerged though.

 

Just as a review, let's start with the good facts and finish with the newly appeared problems:

 

The fixes:

 

1. AoA at or near 90 is now fixed;

2. Realistic pitching moment variation between stall AoA and maximum reachable AoA for a constant elevator deflection, as a function of AoA is obtained.

3. Impressive and reality related pitch-roll oscillations during spins according to elevator inputs and existing yaw rates is now achieved.

 

New issues:

 

1. The rudder effect above stall AoA is quite high. For instance you can induce quite high initial yawing moments during high AoA conditions where most of the rudder would be shadowed by the fuselage and most of the airflow that the rudder would encounter whould most probably be detached/turbulent. During tests, even if you'd allow the mig-21 to gain high yaw rates during flatspins, from the moment you apply full rudder opposite to spin, the plane takes just one or two turns of spin more (which is quite quick) giving the plane's mass and yawing moment of inertia, so either the yawing moment of inertia is low or the yawing moment generated by the rudder is high, but I believe that it is the latter which causes this.

 

2. Ailerons no longer create yawing and rolling moments above stall AoA. This was perfectly modeled before the mentioned fixes were noticed, but now this shows up. It is known that as the AoA increases from null lift towards 90 deg. of AoA, the yawing moment generated by the rudder has a logarithmic decrease towards 90 and remains very low at that angle, the rolling moment due to ailerons also have a logarithmic decrease towards 90 and while the wing reaches 90 AoA the aileron deflections would produce zero rolling moments, except when a yaw rate exists which could allow for slight rolling moments to develop. Also, as the AoA increases between null lift towards 90 deg., the ailerons deflections produce yawing moment which is known as yaw due to rolling input, and this yawing moment should have a logarithmic increase this time. So as the AoA gets higher you would normally find a lower yawing moment from the rudder input but higher yawing moment from the ailerons, these effects reaching their peak at 90 AoA. For some reason, the ailerons no longer produce any roll and/or yaw as the AoA is higher than stall.

 

3. The critical AoA of the MIG-21 has suffered an unexplained decrease from around 20 deg. towards 14..15 deg. In reality the critical AoA will always increase as the aspect ratio is lower, as the wing sweep angle is higher and as the airfoil is thicker and has a higher camber and also as leading and trailing edge devices (including boundary layer control systems) exist. During normal flight without any high lift devices engaged, the critical angle of attack increases for two reasons: higher wing sweep and lower aspect ratio. The MIG-21 has both a very low aspect ratio (almost as low as the F-104 has) and a quite highly swept wing. A B-737 airliner finds a stall AoA at about 14..15 deg. when no slats or flaps are out, but the airliner has a high aspect ratio and lower wing sweep than a MIG-21. The F-15C in DCS finds a stall AoA at around 24..25 AoA and although it also doesn't have leading edge devices it has a similar aspect ratio and sweep to the MIG-21's wing, but even if indeed it has a higher aspect ratio and slightly lower wing sweep than the 21, the difference is dramatic now of about 10 deg. The F-15 indeed has some vortexes generated between the engine inlets and the canopy which energize the boundary layer between the wing's root and fuselage, but they are quite weak and may probably increase the critical AoA by 2..3 deg. The MIG-21 should probably regain it's 20+ deg. stall AoA which was correct.

 

 

Best regards!


Edited by Maverick Su-35S

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

A chart of C-L maxes at different Mach is available in the manual and attached to this post.

 

RE: #3. The MiG-21bis has a built in safety margin where 28 units alpha is equal to 85% of the stall AOA at > 0.85M. For a relatively narrow (but important) Mach range of approx. 0.88-0.95M 33 units alpha by the indicator equates to C-L-max. At low Mach numbers the margin is greatly increased. At high Mach numbers the tail has insufficient authority to reach these AOA.

 

For example at 0.5M 28 alpha represents a mere 70% of stall and 33 alpha about 81%. At these low Mach numbers you should be able to exceed the scale of the gauge significantly before stall occurs. The 33 alpha marks on the UUA-1 are only valid (not premature) stall warnings above about 0.85M.

 

You can make a setup straight out of the book, middle weight, clean, 5000m, level and follow along with the text:

 

A. Prestall buffeting begins at 340-370 kmph and continues roughly the same until stall.

B. Less than 280 kmph ailerons effectiveness considerably deteriorated.

C. At 230-240 kmph rocking in roll as large as +-50 degrees (200 kmph).

D. Reduced to 220-230 kmph, control stick fully back the airplane starts "pancaking" with the oscillations in roll of 6-10 deg/s.

E. The stall airspeed without opposing these osculations is 220-230 kmph at idle thrust.

 

I find A, B, and to an extent E to be modeled in DCS but C and D not. The airplane departs in pitch at 16 AOA (33 alpha) despite the low Mach number. That's not to say this is necessarily incorrect behavior as stability and controllability are practically independent aerodynamic concepts. To evaluate lift and stability in the region between 33 alpha and the Cy-max the airplane must be placed there for a meaningful length of time.

 

The problem for the person who only has DCS as a tool is that it is difficult to explore this alpha range. The increase of AOA is so rapid that even a determined reaction of forward stick cannot capture this regime. Either the reaction is excessive and produces lower alpha or insufficient and the alpha range of interest (33~50 units) is passed.

 

One reason for the difficulty is our inability to operate on the real controls where the combination of stick position and stick forces cannot be realized with common spring joysticks and force feedback devices have been hopelessly broken for years. Our virtual pilot dutifully maintains stick position without giving any hint as to the forces (direction and magnitude) required to achieve it.

 

Using fractional time acceleration I believe I am able to balance the airplane at 18-20 AOA (just slightly beyond 33 units). The result is a 0.2G load factor which is waaaay too little. That AOA and corresponding alpha in this condition should result in a CL which is probably superior to the CL at 33 units indicated being likely very near the peak of the CL graph. In the unlikely event that I'm slightly over CL max I should be no worse than I was at 33 units.

 

I think a more determined effort with more precise loading corrected for the alignment (vertical G not equaling G parallel to gravity due to pitch) and the variable speeds involved could piece together a CL graph but it would be crude and error prone.

 

Overall I believe the developers have coded an overzealous drop off in lift through the peak lift coefficient combined with a naive acceptance of the regulatory UU-1 limits as being the same as the true aerodynamic characteristics. The CL graph should be roughly symmetrical about the maximum point and quite flat. If the peak is at 20 degrees AOA for example the CL at 22 should be about the same as at 18.

 

I really want to see some graph printout from the LNS development widget of CL, at low Mach, vs. incidence angle. They must have some sort of virtual wind tunnel build of DCS where they can create various conditions and record the results without being in flight. I have a feeling the graph would show the floor falling out of lift in a very weird way. It's like the MiG enters the vacuum of space past 33 units and never reaching peak CL which in most cases should happen significantly beyond 33 units.

 

For example if 28 units if 85% of the alpha to CLmax and 33 units is 100% then it is expected roughly that whatever AOA corresponds to 38 unit will fall back down to very roughly 85% of CLmax again. Decreasing to tiny fractions of CLmax (~20%) a couple degrees over the peak is not reasonable.

 

I understand that controllability can be weird and obfuscate things with stalls but provided that you can get the airplane into these situations the results should make sense. Controllability is another topic for another day.

Fig6.gif.f5f148b2116895289de942b99c7d36a5.gif

Link to comment
Share on other sites

For example at 0.5M 28 alpha represents a mere 70% of stall and 33 alpha about 81%. At these low Mach numbers you should be able to exceed the scale of the gauge significantly before stall occurs.

 

It's remarkable that DCS's professional flight model not only take aerodynamic forces and moments as function of AoA and beta, but also with Mach number, covering very realistically the behavior of the real aircraft. The most important changes in lift, drag and moments coefficients for a fighter occur throughout the transonic regime where dramatic changes in aero forces and moments develop between M 0.85 and M 1.0 (as it may be the case for MIG-21) and find a somewhat mirrored recovery towards M 1.2...1.3. Although not shown in that very valuable diagram that you linked, normally from where the critical Mach number starts until Mach 1 is reached, the CL (lift force coef.) starts to constantly drop due to emerging shock stall effects, the CD (drag force coef.) starts an abrupt constant rise and gets about double that of the low Mach incompressible flow which makes it hard to pass through Mach 1, the Cm (pitching moment coef.) constantly goes way more negative and between Mach 1 and Mach 1.2 or 1.3 the variation of these variables is almost a mirror of how they develop between M 0.85 and 1.

 

One reason for the difficulty is our inability to operate on the real controls where the combination of stick position and stick forces cannot be realized with common spring joysticks and force feedback devices have been hopelessly broken for years. Our virtual pilot dutifully maintains stick position without giving any hint as to the forces (direction and magnitude) required to achieve it.

 

I agree that there's a difference between a virtual pilot's spring loaded stick and a real MIG-21 in flight hydraulically loaded stick with feel actuators, but even so the difference shouldn't be so great even without force feedback giving these reaction conditions where even a real 21 pilot couldn't hold an alpha between 15 through 20 and vice-versa as he wants no matter how hard he'd try if the plane would react as in DCS like it slightly jumps through those AoA ranges and it would be even more impossible for the real pilot to do that because his stick has a limited maximum travel speed while even us with a virtual and quicker stick can't hold a constant alpha through that AoA range;).

 

Using fractional time acceleration I believe I am able to balance the airplane at 18-20 AOA (just slightly beyond 33 units). The result is a 0.2G load factor which is waaaay too little. That AOA and corresponding alpha in this condition should result in a CL which is probably superior to the CL at 33 units indicated being likely very near the peak of the CL graph. In the unlikely event that I'm slightly over CL max I should be no worse than I was at 33 units.

 

So you also agree that there is a problem with the G-load drop after the stall (as I've talked about in my first posts here) which no other plane in DCS does so this is abnormal for to the 21 only. Giving real experimental results, at the bottom of the CL vs AoA function beyond sall, the CL should be somewhere between 60% through 100% of the maximum CL found at the stall point depending on wing aspect ratio (ex: the higher the wing aspect ratio and thinner airfoil, the greater the drop), wing sweep and boundary layer control which is either through vortex generators such as LERX or blown trailing edge devices.

 

I think a more determined effort with more precise loading corrected for the alignment (vertical G not equaling G parallel to gravity due to pitch) and the variable speeds involved could piece together a CL graph but it would be crude and error prone.

 

Indeed it's always better to use real experimental data as input when you have it, cause that's why we sometimes find weird aircraft behavior in DCS if the aerodynamic forces are only evaluated using virtual wind tunnels.

 

Overall I believe the developers have coded an overzealous drop off in lift through the peak lift coefficient combined with a naive acceptance of the regulatory UU-1 limits as being the same as the true aerodynamic characteristics. The CL graph should be roughly symmetrical about the maximum point and quite flat. If the peak is at 20 degrees AOA for example the CL at 22 should be about the same as at 18.

 

Right..., or any difference should be little anyway but this is mostly for aircraft with less wing sweep and higher AR(aspect ratio). For lower AR and higher sweep as the MIG-21 has, the generated vortexes at the leading edge are significantly strong to reduce the CL drop slope as AoA goes past beyond stall, so not only that it should be somewhat rather symmetrical, but may even not drop that much anyway. I personally believe (just by experience, I didn't stand to make any experiments) that at the bottom of the drop, the G-load should be at least 80% the maximum achieved at stall.

 

I understand that controllability can be weird and obfuscate things with stalls but provided that you can get the airplane into these situations the results should make sense. Controllability is another topic for another day.

 

As you've said, controllability and static and dynamic stability are not necessarily linked and are mostly individually linked to aerodynamic effects, but it's clear that these effects should be meticulously examined bit by bit for how does the aircraft reacts in accordance with reality when real data is available either if a real test pilot can share it or if experimental data exists. The fact that after newer updates the ailerons no longer give any effect after the plane passes beyond stall AoA, which is indeed a controllability problem while the rudder effect is kind of too great otherwise, is indeed a new aspect to talk about.

 

I'm very glad for these constructive and realism based conversations.

 

 

Best whishes!


Edited by Maverick Su-35S

When you can't prove something with words, let the maths do the talking.

I have an insatiable passion for helping simulated aircraft fly realistically!

Sincerely, your correct flight model simulation advisor!

Link to comment
Share on other sites

I'm not as competent as You guys in discussing these post-critical regimes, but I also found rapid "pancaking" to the ground somewhat eyebrows-rising. While attempting to test stalls and spins behaviour, with stick full back, plane drops like as stone, reaches and exceeds 400 km/h while falling almost flat - would it be possible for the object of this size, at the most drag-inducing attitude (that reminded me supersonic flat spins we saw in the very first version of the module :D)? The entry is so fast that with G's dropping to zero, it's difficult to practice these manoeuvres now without imminent engine flameout (how do You guys do spin entry without stalling the engine?).

 

I also have mixed feelings about not even a tiny wing drop during this "deep stall". While testing from 10 km of altitude down to 0, the plane drops with wings dead-level (I could've sworn that was not the case in early 1.5.4) and starts rolling only about 2 km AGL. The funny thing is, when it rolls sufficiently, a sudden, magic, great lift force out of nowhere saves it from certain crash and shoots it back up - I guess simulation of lift during sideslips goes crazy, but I don't mind, as such a silly test pushes the limits of the game code already - I wouldn't notice it during normal flights anyway. I would, however, like to see the stall entry phase being looked into by Novak.

i7 9700K @ stock speed, single GTX1070, 32 gigs of RAM, TH Warthog, MFG Crosswind, Win10.

Link to comment
Share on other sites

Throwing super rough figures into a terminal velocity calculator of 7.5MT, 50m^2, 0.8kg/m^3 (3km), and a Cd of 1.0 gives an output of 60m/s. I don't think the MiG falls faster than that at 90 AOA. However at terminal the G meter should rise again to 1.0.

 

I wish I had a CFD program, a 3D model of the MiG, and spare enough CPU cycles to make answers out of questions. Somewhere in some filing cabinet is wind tunnel data for this airplane.

Link to comment
Share on other sites

  • 1 month later...
  • 6 months later...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...