Jump to content

The new critical angle of attack might be too low!


Maverick Su-35S
 Share

Recommended Posts

Sorry if I'm starting this thread and a similar one is already there, but I'd rather make this subject fresh as it regards new FM updates done to the MIG-21.

 

In the latest version, the MIG-21's FM won't allow for an angle of attack greater than 15 before the stall onset. Initially the critical angle of attack was fine and very appropriate for this type of wing at about 20 degrees even if it doesn't have leading edge droops. For the low aspect it has and high sweep, 20 deg. of AoA was quite fair.

 

I personally can't understand why a decision was taken to lower the critical AoA to about 14-15 degrees, which is even lower than that of the L-39 (16..17AoA) for instance, or than the A-10 (which has 17 to 18.) or Su-25 (17 AoA). I can't get the logic behind this very low value of critical angle of attack for the 21.

 

If is there any wind tunnel or other hard proof which shows that the MIG-21 in clean config (no external stores) with flaps up starts to stall at 14..15 deg AoA I accept that, although it's very non-logic for this to happen.

 

 

All the best!


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 fanatic!

Link to comment
Share on other sites

  • Replies 167
  • Created
  • Last Reply

Top Posters In This Topic

Because balalaika wing.

https://leatherneck-sim.mantishub.io/view.php?id=452 look karasik posts.

 

In DCS good PFM have ED and Belsimtek, all other 3rd developers shit tier, IMHO.


Edited by Yappo
  • Like 2

CPU: i7-8700K @5200MHz / RAM: G.Skill Trident Z 2x16GB @3500MHz CL15 / MB: ASUS PRIME Z370-A / GPU: ASUS STRIX RTX 2080 / DCS-Storage: Samsung 960Evo 500GB / Thrustmaster Hotas Warthog + T.Flight Rudder Pedals

[sIGPIC][/sIGPIC]

Link to comment
Share on other sites

Cy-max (maximum lift coefficient) should correspond to the 33 alpha gauge indication at a certain Mach value (high subsonic). At lower Mach the peak lift should happen beyond 33.

 

I find DCS has stall approximately where UUA-1 indicator readings end under all conditions of Mach. And really Cy-max should be in excess of 33 alpha below about 0.93M. Unfortunately UUA-1 indication is a poor yardstick because true to UUA-1 reading relationship is not known (at least by me) so it can't be verified.

 

The change in UUA-1 reading (if the scale went that far) of where CL max is possibly due to CL peaking at larger true angles at low Mach or it could be because UUA-1's relationship to true angle is strongly changing based on Mach. Recent test shows DCS UUA-1 is a simple and constant y=mx+b relationship with true angle. Who knows what the real relationship is?

 

No matter what the stall (CL peak) is we expect the CL curve swept over the peak value AOA it should be roughly symmetric before and after peak lift. Calling peak lift 100% if 1° less is 90% then 1° post stall should be ~90% as well. Sufficiently beyond the peak angle the lift dies away more rapidly than it was on the opposite of the peak (symmetry breaks down). So evaluations of peak lift (say by generating load factor) should be identifiable as angle increases the lift (or at least CL) should decrease smoothly post-peak as it rose pre-peak for small angles around the peak.

Link to comment
Share on other sites

Because balalaika wing.

https://leatherneck-sim.mantishub.io/view.php?id=452 look karasik posts.

 

In DCS good PFM have ED and Belsimtek, all other 3rd developers shit tier, IMHO.

 

Sadly, I feel the same at the moment! ED and Belsimtek seem to be the only guys to master an FM and make it realistic or PFM level. Maybe it's not my right to talk about it, but I hope the other third parties will also be able to learn from ED how to work out their FM's to make them PFM. ED should guide/teach them to achieve that, IMO.

 

 

Happy new year!

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 fanatic!

Link to comment
Share on other sites

...

The change in UUA-1 reading (if the scale went that far) of where CL max is possibly due to CL peaking at larger true angles at low Mach or it could be because UUA-1's relationship to true angle is strongly changing based on Mach. Recent test shows DCS UUA-1 is a simple and constant y=mx+b relationship with true angle. Who knows what the real relationship is?

 

No matter what the stall (CL peak) is we expect the CL curve swept over the peak value AOA it should be roughly symmetric before and after peak lift. Calling peak lift 100% if 1° less is 90% then 1° post stall should be ~90% as well. Sufficiently beyond the peak angle the lift dies away more rapidly than it was on the opposite of the peak (symmetry breaks down). So evaluations of peak lift (say by generating load factor) should be identifiable as angle increases the lift (or at least CL) should decrease smoothly post-peak as it rose pre-peak for small angles around the peak.

 

Glad we two have a common understanding about general aerodynamics alone. Indeed the lift drop beyond stall AoA should be nearly symmetrical to how the lift slope developed before it, but even if not necessarily symmetrical (when strong vortexes are generated by the wing root leading edge portion), the lift drop should be very smooth until the vortex actually breaks apart.

 

The NASA F-18's HARV lift/AoA charts and real life aerodynamic tests using tufts is a good source of information:

 

https://www.youtube.com/watch?v=SViiqylV0lA

 

 

Although this topic is about the real MIG-21's critical/stall AoA, the following video is very interesting:

 

 

 

All the best!


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 fanatic!

Link to comment
Share on other sites

Yes infact the MiG-21 atm flies as if its got a FBW AoA limiter making sure the a/c doesn't exceed a certain AoA past ~600 km/h, because above that speed you cannot reach a particularly high AoA anymore. Infact an accelerated stall seems impossible to execute in the MiG-21 below ~600 km/h atm. As a result the aircraft like its on rails in pitch, which is not a particularly authentic flight behavior for a small, pure mechanical control system, delta wing aircraft like the MiG-21.

 

There's also as mentioned a HUGE discrepency between the AoA shown in cockpit versus that on the toggable lower screen information panel, leading me to believe all of this might be due to a mistake in coding.

 

In short the MiG-21 FM is most definitely broken atm, and seriously so.


Edited by Hummingbird
Link to comment
Share on other sites

Hummingbird, the cockpit gauge is from the vane on the fuselage side. The airflow around it is not straight. In no way should the F2 AOA reading and the UUA-1 agree in a direct manner.

 

If you want true AOA from the gauge multiply by 0.51 and subtract 0.72. It works every time. They don't agree and they aren't supposed to.

 

And I don't know what you're talking about I was 600 km/h IAS at 3k with APY in auto and a hard pull pegged the UUA-1 beyond 33.

Link to comment
Share on other sites

Yes infact the MiG-21 atm flies as if its got a FBW AoA limiter making sure the a/c doesn't exceed a certain AoA past ~600 km/h, because above that speed you cannot reach a particularly high AoA anymore.

 

Same as "Frederf" described, you can even go past critical AoA at higher speed if you like and it's normal what the AoA indexer shows.

 

There are 2 reasons why the AoA becomes lower as airspeed occurs though, when pulling the stick:

 

1. There is a hydro-mechanical feedback on the stick which makes the stick stiffer as the aircraft goes faster in order to not let the plane reach critical AoA that easy. This is available for many airplanes, civilian and fighters alike.

 

2. The center of pressure (center of lift acting on the whole aircraft) starts to move towards the rear as the airspeed increases. The faster you go the more rear the CP goes, which makes the angle of attack that you can achieve for the same elevator deflection (if you could hold it) become lower and lower. The slower you go, the opposite happens.

 

These 2 reasons makes the critical angle of attack become harder to reach as the airspeed increases.

 

Try this: Go at a high altitude, pull the stick until you reach critical AoA indicated in cockpit and try to hold it there, then go in full afterburner while descending fast in order to be able to increase speed. You will see that the indicated AoA starts to decrease the faster and faster you go even if your stick might remain in the same position or the elevator deflection hasn't changed. Still, you can stall the wings by trimming as needed then sharply pulling the stick from low initial AoA. This is realistic and normal.

 

Infact an accelerated stall seems impossible to execute in the MiG-21 below ~600 km/h atm. As a result the aircraft like its on rails in pitch, which is not a particularly authentic flight behavior for a small, pure mechanical control system, delta wing aircraft like the MiG-21.

 

You can't do accelerated stalls (going past critical AoA at airspeed that give a lift/weight ratio higher than 1) below 600km/h? You initially said that you can't do this past 600, but neither below?

 

The pitch trim vastly affects the elevator deflection limits by shifting the elevator deflection limits more up or more down when the stick is moved in pitch. So the total elevator deflection is a sum between the deflection given by the stick and the pitch trim. Stick + trim = final elevator deflection.

 

If you trim the pitch to a more nose up position you'll see that you can go past beyond stall AoA at any speed, from 10/km to 700+km/h after you pull full aft stick, especially when doing it sharply form an initial low AoA. Indeed the more you increase speed the lower the achievable AoA becomes, but you can still stall it even higher than 700km/h, not just 600, if you trim the nose quite up then sharpy yank back on the stick, you'll make the plane overshoot the critical AoA for a split moment and cause a wing to stall. This is not a plane on rails in any circumstance and flies realistically.

 

How would you define such a behavior? I would define a plane that flies on rails a plane that doesn't give a correct dynamic in motions, for instance the lack of correct moments of inertia simulation, aerodynamic forces and moments not varying with airspeed in a correct way, scripted flying behavior (programmed to behave in a certain way no matter what, which defies reality), lack of oscillations around axis when full sharp inputs are given the rapidly put to neutral. If one of these happens, you may call it on rails, but so far I personally couldn't find any of these happening in latest patches, so the plane flies like the real one, except for the critical AoA which should be at a higher value, not just 15 as it is now. High sweep delta wings generally have critical AoAs higher than 20 even without slats/droops, so this might be the only problem so far.

 

There's also as mentioned a HUGE discrepency between the AoA shown in cockpit versus that on the toggable lower screen information panel, leading me to believe all of this might be due to a mistake in coding.

 

In short the MiG-21 FM is most definitely broken atm, and seriously so.

 

I've already explained why there's the difference in real AoA and shown AoA and this is a realistic behavior reflecting how the planes were built those days. They didn't care those days that the vane would deflect more than the real AoA at which the plane flies as they didn't care as the corresponding 33 AoA in cockpit corresponds to a real 20 AoA of the aircraft. Now it's 15 only and I bet it's wrong, but anyway, the difference between shown AoA on the cockpit indexer and real AoA is something normal.

 

 

All the best!


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 fanatic!

Link to comment
Share on other sites

Thanks for the corrections on the vane AoA readings guys, I didn't realize that! :thumbup:

 

Still I find it odd that I can't seem to get passed 15 deg true AoA as speed rises above ~600 kmh TAS, despite not touching trim. This didn't use to be a problem at all, and now the aircraft doesn't really seem capable of flight at particularly high AoA anymore.

 

And I don't know what you're talking about I was 600 km/h IAS at 3k with APY in auto and a hard pull pegged the UUA-1 beyond 33.

 

This suprises me a lot, going ~650 km/h TAS at SL I was unable to exceed around 20 deg on the UUA-1 IIRC. It was as if something was limiting the amount of pitch input I was able to produce.

 

You can't do accelerated stalls (going past critical AoA at airspeed that give a lift/weight ratio higher than 1) below 600km/h? You initially said that you can't do this past 600, but neither below?

 

Sorry that was a typo, meant above 600 ofcourse, as below 600 is the only time I seem to be able to reach 30 deg AoA on the indicator.

 

How would you define such a behavior? I would define a plane that flies on rails a plane that doesn't give a correct dynamic in motions, for instance the lack of correct moments of inertia simulation, aerodynamic forces and moments not varying with airspeed in a correct way, scripted flying behavior (programmed to behave in a certain way no matter what, which defies reality), lack of oscillations around axis when full sharp inputs are given the rapidly put to neutral. If one of these happens, you may call it on rails, but so far I personally couldn't find any of these happening in latest patches, so the plane flies like the real one, except for the critical AoA which should be at a higher value, not just 15 as it is now. High sweep delta wings generally have critical AoAs higher than 20 even without slats/droops, so this might be the only problem so far.

 

Well it's most likely the low critical AoA causing this feeling with me when flying it then.

 

Also when I say "on rails in pitch" I am refering to the aircraft achieving a max rate turn at a relatively low AoA with no ability to exceed the critical AoA, the former not being typical delta wing aircraft behavior and the latter not being typical non FBW aircraft behavior.


Edited by Hummingbird
Link to comment
Share on other sites

Ok, retested it today offline, and I must have misremembered the speed or something odd happened online.

 

Anyway what I found was that if I initiated a max performance turn ~750 kmh TAS or above then I couldn't achieve an accelerated stall (AKA reaching the critical AoA) until the speed in the turn gradually dropped to around 600 km/h - the AoA gradually increasing as the speed in the turn dropped despite no change in stick deflection which was max from start to finish. (The aircraft was trimmed for straight flight at 650 km/h)

 

In addition to this the stall happened at around 15-16 deg true AoA as mentioned (at 750 I was only able to pull ~12 deg), which seems extremely low for a delta. Infact the MiG-15 & Sabre both have a higher true critical AoA ingame...


Edited by Hummingbird
Link to comment
Share on other sites

T...

 

Also when I say "on rails in pitch" I am refering to the aircraft achieving a max rate turn at a relatively low AoA with no ability to exceed the critical AoA, the former not being typical delta wing aircraft behavior and the latter not being typical non FBW aircraft behavior.

 

I've got you now. Indeed the lift/AoA slope is higher than normal, at least from the way I see it. The way it's modeled right now seems more like an A-10's lift slope (it isn't necessarily that, but it looks like), just with the difference that at a 0 (zero) angle of attack the lift is close to null as it should.

 

The FBW kind of response felling is more probably related to how the hydro-mechanical elevator deflection is being modeled in DCS in correlation with the flying indicated airspeed (related to dynamic pressure). Only experimental tests could reveal how the elevator deflection would be limited according to the dynamic pressure at which the plane flies.

 

My biggest concern for now is the critical AoA level which seems way too low.

 

 

All the best!

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 fanatic!

Link to comment
Share on other sites

...

 

In addition to this the stall happened at around 15-16 deg true AoA as mentioned (at 750 I was only able to pull ~12 deg), which seems extremely low for a delta. Infact the MiG-15 & Sabre both have a higher true critical AoA ingame...

 

You couldn't be more right!:thumbup: This is the problem that the devs should re-look into, cause I've talked to real MIG-21 pilots and they also know that for the indicated 33 AoA in cockpit, the plane realistically reaches a plus 20 AoA on the wing AND even if passing beyond that the plane may get a roll-off (fast snap roll due to one wing stall) but then quickly recover (roll rate becomes null quite shortly) then continue to be very docile as the AoA continues to increase beyond stall with slight lift loss (didn't say how much). So, the lift loss should also be not as dramatic as it's in-game right now, where the G-load still drops below 50% of the value before the stall. I guess the lift (and of course, G-load alike) should remain somewhere between 80-90% the maximum achievable, not just 50%.

 

For short, we should be looking for a higher critical AoA, lower G/lift drop beyond stall.

 

Let's wish for the best!


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 fanatic!

Link to comment
Share on other sites

This suprises me a lot, going ~650 km/h TAS at SL I was unable to exceed around 20 deg on the UUA-1 IIRC. It was as if something was limiting the amount of pitch input I was able to produce.

 

There's a system that does exactly that! It even has its own gauge below the nose cone position indicator. It changes the ratio between stick and elevator movement as airspeed increases. The gauge shows the current pitch ratio. Know your airplane, guys ;)

 

It does start limiting max AoA below corner speed, which doesn't make sense. You end up not being able to produce max AoA OR G load at certain speeds.

DCS modules are built up to a spec, not down to a schedule.

 

In order to utilize a system to your advantage, you must know how it works.

Link to comment
Share on other sites

There's a system that does exactly that! It even has its own gauge below the nose cone position indicator. It changes the ratio between stick and elevator movement as airspeed increases. The gauge shows the current pitch ratio. Know your airplane, guys ;)

 

It does start limiting max AoA below corner speed, which doesn't make sense. You end up not being able to produce max AoA OR G load at certain speeds.

 

Yeah I was informed of this earlier, however the low critical AoA in combination with premature activation of this system produces an odd feeling when flying the aircraft.

Link to comment
Share on other sites

Wrong. The 1° angle you are talking about is actually zero for MiG-21.

 

LOL! 1 degree? 1 degree equals nothing compared to what happens with the real critical AoA. +/-1 isn't a big deal! 15 AoA instead of 20 is a big deal as it affects the whole lift/AoA derivative (slope). Don't want to contradict you now, but I later found out that it is indeed +1 degree though and "Frederf" was right. I've talked to a real MIG-21 pilot and he told me that the wing has 1 deg. of incidence at root. That guy knows the 21 by the book and also confirmed me that the real critical AoA is around 20 for no flaps and slightly increases with flaps using boundary layer control.

 

Good day!


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 fanatic!

Link to comment
Share on other sites

...

It does start limiting max AoA below corner speed, which doesn't make sense. You end up not being able to produce max AoA OR G load at certain speeds.

 

You mean above, not below.

 

It makes sense that the hydro-mechanical system was built so that the ratio between stick deflection and elevator deflection drops as airspeed increases for the fact that the critical AoA also decreases as airspeed increases since the airflow becomes compressible (above 300km/h). So after +300km/h, the compressibility factor of the air starts increasing rapidly and so the critical AoA that can be achieved drops up to transonic airspeed where a shock stall effect also becomes present for all known airplanes, then the maximum AoA slightly increases back through supersonic up to some value when the Mach number gets 1.2-1.3, then again starts decreasing as airspeed continues to increase. This is the aerodynamic reason why the achievable AoA needs to be reduced as airspeed increases, as a means of making keeping you clear from the critical AoA first of all and secondly to not allow for too high G-loads that can result. Nothing's wrong with the simulation in this area so far.


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 fanatic!

Link to comment
Share on other sites

He means below. If the corner speed is 950 km/h then the APY starts reducing tail deflection at speeds less than that.

 

What should be accounted for is that stick forces vary wildly with airspeed and APY ratio and being away from the auto-scheduled values produce problematic handling both for the pilot and SAU. It is not cruelty to adjust the ratio as if to sadistically inhibit the pilot from achieving maximum rated loading at certain airspeeds. Even at the smallest arm setting the maximum permissible alpha should be attainable at all subsonic speeds.

 

In any case this LanceR demo shows what I think is the same engine, airframe, airfoil, and control system doing comfortably maneuvers in the range of 20-30 true AOA.

 

${1}

Edited by Frederf
Link to comment
Share on other sites

LOL! 1 degree? 1 degree equals nothing compared to what happens with the real critical AoA. +/-1 isn't a big deal!

 

And what exactly are you LOLing about? I only corrected an affirmation made by Frederf about the real aircraft. I didn't say if 1° is a big deal or not.

 

Don't want to contradict you now, but I later found out that it is indeed +1 degree though and "Frederf" was right. I've talked to a real MIG-21 pilot and he told me that the wing has 1 deg. of incidence at root.

 

That's nice but I do want to contradict you right now. The attached image is a page from "MiG-21bis flight-technical chracteristics" manual.

21wing.thumb.jpg.01b90eaea816e54d1c4f717722b16e78.jpg

Link to comment
Share on other sites

...

That's nice but I do want to contradict you right now. The attached image is a page from "MiG-21bis flight-technical chracteristics" manual.

 

Maybe the one that I heard to have 1 degree might be a different custom version, idk., cause I can't yet find an example either of a MIG-21 having a wing root incidence different than 0 on the internet right now, but possibly there might be one, yet I just didn't find it very important that the wing might have 1 degree of incidence or zero. That difference isn't of much importance. There are other issues that are important which concern the topic here and other things like the lift slope of the MIG-21 and zero lift angle of attack of it, which are strange.

 

 

Good day, and sorry for being too quick on the LOL by thinking that you were wrong in fact, cause it seems that I can't find a different example about it either, so I can't contradict you about the wing root incidence.

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 fanatic!

Link to comment
Share on other sites

Stupid question incoming...

 

The designers of the MiG used the fuselage mounted AoA probe knowing it wouldn't be accurate.

 

Why not use the AoA/sideslip vanes on the pitot tube? Wouldn't they give an accurate reading due to them being in clean air? What are they currently used for?

DCS modules are built up to a spec, not down to a schedule.

 

In order to utilize a system to your advantage, you must know how it works.

Link to comment
Share on other sites

The gun sight.

 

The fuselage-mounted vane might be more representative of the airflow local to the leading edge of the wing (which would be preferable to the overall angle given the choice). It doesn't really matter if it's accurate as long as it is more consistent so that rules can be formulated for it. There might also be some interface with other systems that the DUA-3 was just better (or not incompatible) with the PWD-7 boom. The bis series having specific boom vane improvements for the purposes of air-to-surface aiming might have precluded using that sensor source also for general flight.

 

The MiG-21M models and later included the side vane.

Link to comment
Share on other sites

Pocket Sized, as had Frederf said, the boom vanes are used for the gun sight.

The side fuselage vane came as a part of the autopilot package.

 

As for which is more accurate... you are thinking correctly that boom readings should be closer to the real AoA. However boom installation has other problems which fuselage does not. Like oscillation of the boom under G, etc.

So, there is no silver bullet... you need calibration for all kind of installations.

Link to comment
Share on other sites

 Share

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...