Jump to content

The new critical angle of attack might be too low!


Maverick Su-35S

Recommended Posts

Hi,

 

It seems that purely accidently I have found the answer to the question: How are the true (not local UUA-1 angle ) AOA for the MiG-21. I found a guy who sells different parts of the MiG-21. One of those things is:

 

e79a9d6cb844756b.jpg

 

7038a1c4f3636951.jpg

 

f0dcf70bc10f5831.jpg

 

7166e101c6a8dde9.jpg

 

 

 

It's a device used to calibrate the DUA-3 sensor , inside the storage packaging we have sticker, with interesting information on it.

 

83e8440d1b69ae39.jpg

 

The main inscription reads: "Deviation angles of the vanes corresponding alpha and beta angles for duas", first box "protractor indications", and then "alpha angles" and "beta angles." I understand that up to 12º48´ indications are non-linear and then linear with a constant ratio. If my way of thinking is correct, this means that above 12º48´ UUA-1 relation to the true AOA is fixed at 1.28 (counted after conversion from seconds to hundredths). So 33º UUA-1 critical angle ( stall limit ) is 25,8º true AOA and the maximum safe angle of 28º UUA-1 is the 21,9º true AOA. Now we have in module about 16-17º true AOA.

Link to comment
Share on other sites

@foxbat155

 

The device in your pictures is for calibrating the DUAS-69 probe (the alpha/beta probe on the Pitot boom), as you can see from its name DUAS69 and the alpha/beta rows in the table. DUAS comes from datchik ugla ataki i skoljenia - that is alpha/beta probe.

 

The probe on the side of the air intake feeding cockpit AOA indicator is called DUA-3.

Link to comment
Share on other sites

  • 4 weeks later...
  • 3 weeks later...

Wow! So tweaking aircraft's reactions isn't the problem, the problem is getting the right values correctly.

 

With the latest update: https://forums.eagle.ru/showthread.php?t=147601&page=5, the MIG-21 has somewhat closer to real aerodynamic moments in relation to the moments of inertia which are responsible for the pitch, roll and yaw accelerations, but, now they're exaggerated.

 

It's a step closer to real than what it was before, but now on the other extreme!

 

Although I had started this topic about the critical AoA value being too low (and it still persists that way, because instead of about +20, we have 15) and the fact that the lift vs AoA slope is too steep and the null AoA lift is too high for MIG-21, I slide for a bit to also talk about the angular accelerations about the 3 main axis which are also a concern.

 

About 2 months ago I posted 2 video links, where the first one shows exactly the time (can be calculated) needed for the MIG-21 to accelerate in roll from 0 to maximum roll rate when flying at about 500 km/h and weighing around 7500kg (50% fuel) within an air density of around 1.12, from where an initial roll acceleration can be derived. The calculated rolling acceleration can be used either to estimate the correct rolling moment of inertia or the output aerodynamic rolling moment.

 

Now after the latest update the MIG-21 seems to accelerate in roll (although pitch and yaw are also affected) at about 200..250km/h with the same amount of rolling acceleration as the real aircraft at 500km/h. Keep in mind that all the aerodynamic forces and moments vary with the square of the airspeed, so you can estimate/calculate how great the accelerations about all 3 axis are at 500km/h if the required amount for that speed is already there at about just 230km/h.

 

The good thing is that now the general aircraft motions or dynamics (especially in pitch and roll) are better (much closer to real) and we don't see those twitchy/sudden and very high pitch and roll moments/motions anymore above stall AoA, which makes the aircraft feel again like something real. So now the airplane's responses are as smoother as they should, but the accelerations are too great.

 

Here are the good and bad that can be found after the latest update:

 

The good:

 

-Smoother dynamics/motions of the aircraft about the 3 axis between -180 and +180 AoA and beta angles.

 

-G or lift above stall AoA is now higher than before (about 0.6 times that at stall AoA), but still not high enough to be realistic. Take for example the A-10 or Su-25 or F-15 and compare how much their lift drops well above stall in relation to that of the MIG-21. You'll find out that the MIG-21 still drops like a rock well above stall, so it still needs refine. I'd guess that the 21's wings should still produce about 80%..85% of the maximum lift some 10 degrees above stall AoA and not just 55%..60% as it is right now.

 

-Other FM fixes.

 

The wrong:

 

-Although it's very good that the ratio of aerodynamic moments to inertial moments is higher for "X" airspeed, it now became exaggeratedly high and must be lowered by some amount to become realistic.

 

-The lift drop above alpha stall is still sharp. Just 0.5 degrees of AoA above stall are needed to make the lift drop instantly from maximum to minimum. The lift above stall AoA should have a curved pattern that is generally spread between 5 (for low sweep & high aspect ratio wings) and 10 (for high sweep & low aspect ratio wings) to almost 50 degrees of AoA (for wings covered by high energy vortexes, such as those generated by LERX on various fighters like F-18/Su-27/F-16/MIG-29).

 

The MIG-21's wings in reality are capable of allowing around 10..12 degrees of AoA (from my aerodynamics experience) over which the stall pattern should spread until the lift finally drops to 80% that of maximum. For short, as the MIG-21's wings should start stalling beyond 20 degrees AoA, the lift should gradually drop (with quite a smooth curved function) from 20 until 30..32 AoA is reached and above 30..32 AoA as the alpha continues to increase, the lift should again start to increase with an initial slope which is about half the normal (non-stalled) slope through a secondary curved lift to AoA function up to about 45 degrees AoA, from where the lift should finally start dropping with a cosine function of AoA until reaching zero at exactly 90 AoA. That is how a correct lift pattern versus AoA looks like for a real 3D wing.

 

I provided a track showing how easily you can do power-loops (kulbits) one after another with MIG-21 or obtain some insane instant yaw rates and beta (sideslip) angles when jerking the rudder, either due to it's too low moments of inertia or due to too high aerodynamic moments:

 

MIG-21's new aerodynamic moments too high or inertial moments too low.trk

 

Here's an illustration of how the lift should generally vary with AoA from -180 to +180 for a high sweep delta like the 21's:

 

2030669665_MIG-21aero.thumb.jpg.320ac48bddc54101cb41e4b872bd6956.jpg

 

Here are some real lifting performance info for the 21:

 

312116357_Cz-Mach.thumb.JPG.26ea565772819e2b22862ed469bf30cd.JPG

933211627_F-14vsF-4vsMIG-21.thumb.jpg.ad02dfdd0de4bfe24408bbd4e36305fa.jpg

 

I'm sorry I turned the topic regarding the abnormally low critical AoA alone into a more detailed subject, yet I hope I don't have to start 2 more topics for linked subjects regarding lift, AoA and moments of inertia. I wish that the devs will have a look at it and keep tweaking the values until they get right. I don't know how the mathematical model used by ED/Leatherneck works to get lift and drag vs AoA values, but if this aircraft is to become realistic, it's lift, drag and critical AoAs must get as close as possible to the values given by available charts.

 

Although I may be a critic throughout all my forum discussions regarding how aircraft behave in DCS, I'm only doing it to help get them better in areas where they should normally get better, otherwise I could simply not care and leave all sorts of abnormal things neglected and lie to myself that I'm playing a realistic flight sim. That is simply not what I desire!

 

 

Best wishes!


Edited by Maverick Su-35S
  • Like 2

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

Interesting about the double-humped lift curve (big hump at stall, small hump ~45°).

 

I still wish I had a series of true vs. gauge AOA curves for various Mach graph (say 0.5 through 2.0 in 0.1 increment). The mx+b relationship at all Mach we have now seems too simple to be true and we could learn so much from them.

Link to comment
Share on other sites

Just gonna leave this here.

[YouTube]CaHJLdc20Bc

 

Joking aside, I really like the reduced dynamic pitch stability (I think that's what its called), makes it much more challenging to precisely control AoA when landing and maneuvering.

 

Also, I recall there being some controversy surrounding the stabilization mode of the SAU. I wonder if that is going to be improved in the near future?

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

...

I still wish I had a series of true vs. gauge AOA curves for various Mach graph (say 0.5 through 2.0 in 0.1 increment). The mx+b relationship at all Mach we have now seems too simple to be true and we could learn so much from them.

 

You are perfectly right my friend. If someone would look into how the aerodynamic model (lift, drag, pitching moment vs AoA, Beta and Mach) is done will see that it's very simplistic, very few curved functions, mostly constant values (at least with Mach) or simple linear functions. I don't want to spark disputes by talking about other simulation names, but there are simulators which are highly advanced in flight mechanics and aerodynamics and DCS is kind of falling behind if not taking actions for improvement.

 

The picture which shows how the aerodynamic values vary from +180 to -180 is one of those sims and was here long before DCS.

 

...

Interesting about the double-humped lift curve (big hump at stall, small hump ~45°).

 

Yeah, that's because the high sweep delta (with a 2.2 AR (aspect ratio)) generates some relatively strong vortexes (in comparison to a non-LERX lower sweep wing) near the wing's root which has the energy to smooth out or reduce the flow separation in areas closer and closer to the root. This way, beyond a stall angle of 20 AoA (as the MIG-21 should re-find in DCS, which it had when it first appeared), the lift should gradually drop in a curved pattern up to 30..32 AoA (bottom of stalled lift where the flow separation reaches 100% of the chords length as the vortex broke-up), then start rising again naturally (with about half of the non-stalled lift slope) up to 40..45 AoA and because the flow is separated already and the vortex is already dissipated as the lift increases towards 45 you'll get a smaller hump than that for stall.

 

Here's how a LERX affects the lift/AoA slope:

 

https://s23.postimg.org/jhiphb4h7/LERX.jpg

 

The double hump lift curve to AoA is normal for all known airfoils (infinite span wings) and 3D wings, the only differences are indeed the points where the humps lie on the diagram and their curvatures from one wing/airfoil to another.

 

The F-18 for example has a single big hump on the lift/AoA graph somewhere at 40..50 AoA (F/A-18's critical AoA). It's lift to AoA function becomes curved between 30 and 40..50 AoA (from 30 AoA the vortex near the root starts to brake down and the flow starts to separate, that's why the function becomes curved) and between 40 AoA to +70 AoA the lift to AoA function finds a moderate drop which is almost linear. Between 70 and 90 AoA the lift starts dropping rapidly to 0.

 

The F-16 has 2 humps also, the one at the stall and vortex core breakdown (35AoA) is much greater than that at 40..45 AoA, so again you'll have a higher radius lift/AoA curve at stall than at 40..45 alpha. The Su-27 (although I haven't seen any aerodynamic charts anywhere yet) may also have a single bigger hump on the graph or anyway a quite large one before the 2nd and commonly rounded one at 40..45.

 

Here's a crude example of lift to AoA function between +180 and -180 AoA:

 

http://www.aerospaceweb.org/question/aerodynamics/systems/cl-cn.gif

 

Here's the F-18's lift/AoA graph:

 

http://www.rollinghillsresearch.com/Water_Tunnels/F18%20tests.html

 

Here's the F-16's lift/AoA slope:

 

https://s16.postimg.org/eokb6b96t/F-16_Ao_A_CL.png

 

The real MIG-21's lift/AoA diagram should be similar to what I posted.

 

 

Best wishes.

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

Just gonna leave this here.

 

Joking aside, I really like the reduced dynamic pitch stability (I think that's what its called), makes it much more challenging to precisely control AoA when landing and maneuvering.

 

Also, I recall there being some controversy surrounding the stabilization mode of the SAU. I wonder if that is going to be improved in the near future?

 

They either exaggerated the aerodynamic moments (too high) or the moments of inertia (too low), but anyway, I believe they'll quickly fix this one. The harder part to fix is to re-arrange the lift to AoA functions in order to fly a realistic MIG-21.

 

The SAU indeed is affected. I don't know if they modified the SAU also or simply does the aero/inertial moments ratio affects it's behavior, but now the plane finds an induced pitch oscillation form the SAU system.

 

The track I provided shows that you can do the types of powerloops shown on your video (good video showing it) even in straight line, one after the other, so the aero to inertial moments ratio is very high.

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

Although the changes weren't right, we must admit that Leatherneck at least tries to fix the flight model of their product and won't disappoint us on long term.

 

I only wish they would take a look at the charts provided and try to re-model the lift and drag vs AoA curves to match the real ones.

 

The chart called "MIG-21 aero" that I posted is as close as possible to the real MIG-21's aerodynamic performance.

 

If I could only have access to the MIG-21's flight model code, I would patiently correct the default values with the ones in that chart and also calculate/determine the correct moments of inertia. Only then I will really feel that I'm flying the MIG-21.

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

Why do you think they are some "values"? :D

[sIGPIC][/sIGPIC]

I5 4670k, 32GB, GTX 1070, Thrustmaster TFRP, G940 Throttle extremely modded with Bodnar 0836X and Bu0836A,

Warthog Joystick with F-18 grip, Oculus Rift S - Almost all is made from gifts from friends, the most expensive parts at least

Link to comment
Share on other sites

Why do you think they are some "values"? :D

 

Please show me the paragraph that you refer to so we can discuss it. If I provided some aerodynamic data (some means nothing else than real world and reliable data) I did so with the purpose to help fix what is off course.

 

Best wishes!

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

Maverick would be a nice if you could make a report on Mantis about that, cause M3 is pretty quiet on this.

And a question regarding F2 view vs TacView. In F2 view the values shown are 16 AoA post stall while in TacView its about 20-21. But vs other aircraft in F2 view AoA reads are normal, can exceed 20 AoA reading from F2 view, same mirage can exceed 20+++ AoA.

"These are not the bugs you are looking for..":pilotfly:

 

[sIGPIC][/sIGPIC]

 

My YouTube channel

 

SPECS

-AMD FX8370 8 Core Processor 4.2 ghz

-GIGABYTE 970A-UD3P

-GTX 1050 TI Windforce 4g

-16 GB RAM

-Saitek X 52

-FaceNOIRtrack - 3 point clip Red Led

Link to comment
Share on other sites

Please show me the paragraph that you refer to so we can discuss it. If I provided some aerodynamic data (some means nothing else than real world and reliable data) I did so with the purpose to help fix what is off course.

 

Best wishes!

 

 

I think you misunderstood me. I meant why do you think they are just some values to fix? From what I understood the FM is quite "magical" (joking) and for example a value (rate of climb might be the result of some computational formulas based on a lot of input numbers and very sensible to "unpredictable variables... hence the strange bugs we get when one thing is fixed and 2 broken :) . So it could very well be a formula that is wrong totally or partially not a value (number for it).

[sIGPIC][/sIGPIC]

I5 4670k, 32GB, GTX 1070, Thrustmaster TFRP, G940 Throttle extremely modded with Bodnar 0836X and Bu0836A,

Warthog Joystick with F-18 grip, Oculus Rift S - Almost all is made from gifts from friends, the most expensive parts at least

Link to comment
Share on other sites

I think you misunderstood me. I meant why do you think they are just some values to fix? From what I understood the FM is quite "magical" (joking) and for example a value (rate of climb might be the result of some computational formulas based on a lot of input numbers and very sensible to "unpredictable variables... hence the strange bugs we get when one thing is fixed and 2 broken :) . So it could very well be a formula that is wrong totally or partially not a value (number for it).

 

I got you now friend and sorry I misunderstood your first statement! You're right about the computational results which indeed emerge depending on formulas used and that there's a complex mechanism between many formulas which give a final result such as: lift, drag, moments, etc. But, the input to those formulas can be (and most certainly are momentarily) the key to control the output values that the aircraft presents in flight. That's what I'm trying to get to. The key to success is part inputs and part mathematical model used, so if one isn't too accurate enough, the other one has to compensate.

 

Only ED who created the aerodynamic model may have control over it and rather suggest Leatherneck (and other 3rd parties) what must be changed in order to correct flight behavior and increase realism, otherwise it's up to the 3rd parties to adjust the input values in order to obtain the right result with the given model.

 

I have done quite some research on the MIG-21's aerodynamic lift, drag and their functions to AoA alone and by comparing the results with the ones shown by the MIG-21 in DCS, I discovered great differences. For instance, the lift/AoA (one of the most important) slope for the MIG-21 in DCS is almost double (1.81 times higher) that of the real jet. In DCS, for subsonic flights, the lift/AoA (in radians) slope is approx. 3.7, when it should be 2.6. There also seems to be no difference in lift/AoA slope with Mach number either. In reality all the aerodynamic coefficients and their derivatives vary drastically with Mach and Reynolds number. Also the maximum lift coefficient (or maximum lift) is almost 60% higher (1.6 times) than that of a real MIG-21, which drastically affects the realism of turning performance, correct landing approach speeds or flight performance all in one. For instance, the real life experimental data shows that the maximum CL (lift coefficient) for MIG-21 is near 0.9 for a clean configuration (no ordnance, no flaps, no BL control). Only with BL and full flaps it rises to about +1.25. In DCS however, the maximum lift coefficient at currently 15 AoA is found to be around 1.43 just in clean configuration (no flaps, no BL)! I didn't even want to further test how high it gets with full flaps and BL activated.

 

I don't know where to discuss it (maybe another thread regarding just this or still here), but I find it extremely important if we need to feel like flying the real MIG-21 and not some video game.

 

ED was created with this in mind: realistic in-flight aircraft behavior. So..., that's what we're all looking for.

 

All ED's aircraft have more than 95% realism proven (with the existence of AFM and PFM). The 3rd party members however vary drastically at this aspect. Belsimtek "somehow" proved to have aircraft modeled as realistic as ED! How is it that they could do it right the first time and Leatherneck still has to do some work to it? The MIG-21's FM still needs work so far. I didn't test the Viggen yet although I also bought it.

 

It's almost obvious that it's up to each one's interpretation of DCS's flight model and thus, the input values to be used.

 

Sorry for the long talk, but, it's important to address these aspects.

 

 

Best wishes!


Edited by Maverick Su-35S
Added proof values regarding the lift performance

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

Maverick would be a nice if you could make a report on Mantis about that, cause M3 is pretty quiet on this.

And a question regarding F2 view vs TacView. In F2 view the values shown are 16 AoA post stall while in TacView its about 20-21. But vs other aircraft in F2 view AoA reads are normal, can exceed 20 AoA reading from F2 view, same mirage can exceed 20+++ AoA.

 

So you're saying that there is a contradiction between what the TacView's AoA says and what the F2 view's AoA says? If that's the case, then I'm pretty sure the TacView is lying because if you take the MIG-21 and constantly (but slowly) decrease the airspeed while trying to maintain 1G at a vertical speed as close to zero as possible, you'll notice that the plane's pitch attitude angle is exactly +15 degrees when the AoA indexer in cockpit reaches 33 units (as you have zero vertical speed in horizontal flight, the pitch attitude angle and the AoA matches), so TacView is almost certainly not telling the truth and the F2 view tells the truth.

 

This makes me lose some trust in TacView's data.

 

 

Best wishes!

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

Good job Leatherneck! Great work and progress on the 21!

 

After the last patches it seems that the relation between aerodynamic moments and inertial moments has passed through step by step enhancements.

 

Now the pitch and yaw aerodynamic moments to inertial moments ratio seems pretty authentic. Even without maths to prove it, the pitch and yaw stability derivatives and aero moments look very right.

 

The maximum roll rate and aerodynamic rolling moments to inertial rolling moments are also realistic/authentic now, but, there's still some last work to be done: the variation of aerodynamic rolling moments with AoA. So far, no matter the AoA, the aerodynamic rolling moments seem to have the same value from null lift AoA to critical AoA, which still isn't right. The rolling moments and roll rates should drop exponentially from the highest value (found near zero lift AoA) to the lowest value at stall AoA. Above stall, the rolling moments should be very low and become zero once the beta (sideslip) angle reaches a certain amount combined with the AoA. Even so with these aspects still requiring attention, we saw efforts and great implication from the guys at LN regarding the authenticity of the MIG-21s FM, we're almost certain they'll carry this one out as well. In the near future, the MIG-21's critical angle of attack will also be brought back to +20 (as it was when MIG-21 first appeared in DCS), lower and correct lift to AoA slope and lower/correct maximum lift coefficient. That's all we ask for;).

 

Again, with honest respect and gratitude, good job LN for not letting this baby down! Waiting for the rest!

 

 

Best wishes!


Edited by Maverick Su-35S
Clarified the rolling moments and rates vs AoA.

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 you're saying that there is a contradiction between what the TacView's AoA says and what the F2 view's AoA says? If that's the case, then I'm pretty sure the TacView is lying because if you take the MIG-21 and constantly (but slowly) decrease the airspeed while trying to maintain 1G at a vertical speed as close to zero as possible, you'll notice that the plane's pitch attitude angle is exactly +15 degrees when the AoA indexer in cockpit reaches 33 units (as you have zero vertical speed in horizontal flight, the pitch attitude angle and the AoA matches), so TacView is almost certainly not telling the truth and the F2 view tells the truth.

 

This makes me lose some trust in TacView's data.

 

 

Best wishes!

 

Yes exactly but the AoA reads on F2 view are always maximum AoA achievable 16 post stall. No matter the speed or flaps usage whilst in F2 view other ac's lets say F-5 can read 20++ AoA. Su-25 has show max AoA 15 before stall. I'm asking is the mig 5 degrees and more less capable on AoA? It also feels it cannot pull anymore as it looks it should.

 

EDIT: Just read your last post which answers my q.

"These are not the bugs you are looking for..":pilotfly:

 

[sIGPIC][/sIGPIC]

 

My YouTube channel

 

SPECS

-AMD FX8370 8 Core Processor 4.2 ghz

-GIGABYTE 970A-UD3P

-GTX 1050 TI Windforce 4g

-16 GB RAM

-Saitek X 52

-FaceNOIRtrack - 3 point clip Red Led

Link to comment
Share on other sites

I think Tacview is showing angle between pitch and flight path while F2 is showing AOA with respect to the air mass. For example while taxing you can get 180 AOA by taxiing in a tailwind while tacview would never know that. I would trust the in-game value more than tacview.

 

In normal few-hundred kmph level flight the difference should be small.

Link to comment
Share on other sites

...No matter the speed or flaps usage whilst in F2 view other ac's lets say F-5 can read 20++ AoA. Su-25 has show max AoA 15 before stall. I'm asking is the mig 5 degrees and more less capable on AoA? It also feels it cannot pull anymore as it looks it should...

 

Speaking of witch, I've read somewhere about the F-5E's in flight performance and was amazed to notice that although it's wings start stalling above 25 true AoA when droops are fully deployed, the flow that remains attached to the wing surface along the path of the strong vortexes created by the LERX (leading edge root extension) only separates above +70 AoA where the vortexes break up (supercritical AoA). Although the F-5's wings have a higher wing aspect and lower wing sweep than those of a MIG-21, which mandatorily reduces the critical AoA, the early F-5's wings which were fitted directly to the fuselage without any LERX or apexes were stalling above 18..19 AoA when the droops and flaps were fully retracted. The droops usually increase the AoA with around 4..7 AoA depending on their design. Slats (fowler leading edge devices) however can increase the AoA with as much as 8..12 AoA (737s slats do so). So even like that (without droops and flaps) the F-5's wings had higher AoA than what those on the MIG-21 right now.

 

The Su-25s indeed start to encounter aerodynamic buffet when the AoA indexer reaches critical and the true AoA is around 15, but the lift to AoA slope is still positive (although curved) between 15 and 18 AoA. Only above 18 AoA the wings of the Su-25s physically start to develop stall (flow separation). The Su-25s, A-10s and L-39s with their straight and high aspect ratio wings still have a critical AoA above 17..18 AoA, so it makes no sense for the 21's wings to stall even earlier than that. The maximum thickness and camber of airfoils indeed affect the critical AoA (higher thickness and camber give higher critical AoA), but only by 2..4 degrees of AoA (so not that much) between a very thin and straight and very thick and cambered one. The strong impact on critical AoA is controlled by wing aspect ratio and sweep. Although both the airfoil shapes and wing aspect ratio and sweep (combined) govern the maximum achievable critical AoA, the airfoils used only affect about 25..30% while the aspect ratio and sweep affect 70..75% of it. So, once more, through every example the MIG-21's wings should provide a much higher AoA before the flow separates.

 

I'm not making this up nor wanting to waste my time, but what I'm saying is based on years of research/experience and can be found on the internet (if one has the proper patience and knows what to search) and technical reports and LN or other 3rd party members can start researching to make sure that people are right and not trolling or trying to create confusion.

 

Here's an interesting article regarding the MIG-21's in flight performance:

 

http://www.military-quotes.com/forum/fighter-performance-actual-plane-analysis-t86206.html

 

Exactly what I'm saying about the critical AoA is also stated here. Just scroll down to about half of the discussion and you'll find this:

 

"Aircraft’s stall speed (speed at which dynamic directional stability breakdown occurs) is function of Mach number, because directional and lateral static stability usually decreases with speed. Stall angle of attack decreases from above 30º (far beyond indicated α) at Mach 0.2 to 20º (i.e. 33 units local angle of attack on indicator) at Mach 0.95.

 

In those days when MiG-21 was designed, electronic flight controls to limit the angle of attack in function of Mach number didn’t exist. A fighter was built primarily for high speeds, high altitude interceptions. At slower speeds previous generations MiG-19/17 were better.

 

Designers put the angle of attack indicator, calibrated in local angle of attack, to warn the pilot of approaching stall limit. At recommended and allowed limit 28 units (about 17º true angle of attack) safety margin to stall is from 13º at Mach 0.2 to 3º at Mach 0.95."

 

"Just before stall α, aircraft nose would start wandering accompanied by more noticeable wing rocking (roll oscillations that intensify thru the stall), symptoms of dynamic directional instability.

Stalling proceeds more vigorously with fewer signs at higher subsonic speeds."

 

The statements tell important (and nice) effects that need to be considered for more accurate/correct simulation. The fact that near stall AoA (18.20) the MIG-21 should experience reduced lateral (roll) and directional (yaw) stability which allows the wings to start rocking more and more as the AoA passes stall (above 20). The rocking effects will normally dissipate a couple of more degrees of AoA above stall (some 2..3 more) while the lift would still remain high (perhaps with slight drops). If not asking too much, we're expecting to see this simulated as well (if possible) in the future updates.

 

 

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

  • 4 months later...
  • 4 months later...

Again..., I don't even know what to do to convince these guys who made the MIG-21 in DCS that THE SIMULATED CRITICAL ANGLE OF ATTACK IS TOO LOW FOR THIS PLANE'S WINGS! All the facts are gathered around and still they won't review this problem?

 

There are 3 modifications that need to be done inside the aerodynamic input data tables:

 

1. Reduce the CL0 - the AoA 0 lift coefficient. It is too high and for this reason this aircraft would be flying supersonic (through shockwaves) at a negative AoA at 1G, which is only possible if this plane has a highly cambered airfoil section, which is not true! The MIG-21 has an almost symmetrical wing airfoil from root to tip (Tsagi S12 has a very low camber), therefore, the CL0 must be very low (quite close to zero). Some data on the internet says it is in fact purely symmetrical, but actual airfoil coordinates show that it is very lightly cambered. Still, the CL0 value must be very low as compared to what it is right now.

 

2. Increase the critical AoA from 15 to 20..20.5, which corresponds to this wing, NOT airfoil, but whole wing.

 

3. Reduce the lift versus AoA slope according such that the maximum CL (or lift coefficient) would not be altered after correcting the critical AoA.

 

3 steps and everyone's happy again. What makes it so hard? You don't trust what I'm saying, right? You need "proof" as always. OK! But one thing's for sure: Even if we don't have access to real wind tunnel data, we should at least agree to the following logic:

 

If the A-10's higher AR wing, zero wing sweep, very high camber as compared to MIG-21's wing, has a critical AoA of 16, the MIG-15, a higher AR, relative sweep, non delta, has a critical AoA of 18, the Su-25, higher AR, very low sweep, has a critical AoA of around 17, how come someone forced a critical AoA on the MIG-21's wing to just 15? Keep in mind that between 2 wings with the same AR and sweep for example, one being a delta and the other a non-delta, the delta wing will always have a higher critical alpha if the airfoils don't differ too drastically (ie: if one is being symmetrical and the other highly cambered)!

 

Just by these contradictions one should understand that even if the true critical AoA isn't exactly known, it should definitely be higher than those of the mentioned aircraft. This is common sense for anyone with a good knowledge and experience in aerodynamics.


Edited by Maverick Su-35S
  • Like 1

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

May I suggest you post these in their Bug Tracker? I really feel your disappointment maybe a post there will trigger them?

 

https://leatherneck-sim.mantishub.io

"These are not the bugs you are looking for..":pilotfly:

 

[sIGPIC][/sIGPIC]

 

My YouTube channel

 

SPECS

-AMD FX8370 8 Core Processor 4.2 ghz

-GIGABYTE 970A-UD3P

-GTX 1050 TI Windforce 4g

-16 GB RAM

-Saitek X 52

-FaceNOIRtrack - 3 point clip Red Led

Link to comment
Share on other sites

I think there is a slight angle between the fuselage datum axis and the wing chord line. And so the "AOA" shown in the software is the angle between the fuselage datum and the air flow. The AOA between the chord line and air flow might have a small fixed offset (~1°?) by construction. It should not be totally assumed that "F2" info bar AOA is showing actual chord AOA.

 

Everything else makes sense.

Link to comment
Share on other sites

  • Recently Browsing   0 members

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