The critical angle of attack is wrong, and its not because the F2 view shows the pitch angle.

[Koty]

2 hours ago, nighthawk2174 said:

Not to mention that if your whole aoa timeframe is compressed by a factor of two small changes in aoa become much harder.  Making the jet harder to fly in general.

Just change your curves. Easy. 21 is as simple to fly as they get.

 

Now, to explain how the UUA-1 works and why it "misreads":

As you might know, the UUA vane is directly on the fuselage, with AoA the flow bends around the fuselage. This is why it reads more than actual AoA really is. Which is why the charts use this indicated AoA as reference rather than real AoA (in the quote in op described as "pitch difference", or difference between aircraft axis and flight path).

If you know the chart reads up to 33° indicated, and you know the UUA is indicating roughly double of actual AoA, then the actual AoA comes out as?

Not to mention stall isn't the only thing that makes aircraft fall out of the sky.

[KenobiOrder]

2 hours ago, Koty said:

Just change your curves. Easy. 21 is as simple to fly as they get.

 

Now, to explain how the UUA-1 works and why it "misreads":

As you might know, the UUA vane is directly on the fuselage, with AoA the flow bends around the fuselage. This is why it reads more than actual AoA really is. Which is why the charts use this indicated AoA as reference rather than real AoA (in the quote in op described as "pitch difference", or difference between aircraft axis and flight path).

If you know the chart reads up to 33° indicated, and you know the UUA is indicating roughly double of actual AoA, then the actual AoA comes out as?

Not to mention stall isn't the only thing that makes aircraft fall out of the sky.

This presumes that the UUA is off by a factor of two, and also that the game models this. I see no reason to make that assumption. The actual stall angle of attack should correspond to the AoA in the F2 view, and right now that is about 15 degrees. Delta wings do not stall at 15 degrees. 15 degrees critical AoA is what you would expect from a tapered straight wing, like on a P-51. Delta wings have a much lower slope of Cl/AoA, and they have lower max CL generally, but they dont critically stall at 15 degees. 

In any case, this is not what the developer stated is happening. They stated that the reason the UUA is different is because they claimed the F2 view is NOT the actual AoA, but is actually pitch angle. Which is why I posted the image at the beginning so everyone would at least theoretically be on the same page as to what that means. AoA is the angle between the flight path and the chord line, and the pitch angle is something else entirely that has nothing to do with fuselage airflow instrumentation errors. 

 

Edited May 23, 2022 by KenobiOrder

[Koty]

Let me make this very simple:
The quote you posted originally talks about a difference between LOCAL AoA, which is angle between aircraft axis and the LOCAL AIRFLOW (not flight path) - and AoA from the F2 view, which is in fact a difference between your PITCH and your FLIGHT PATH, or rather how much below your nose is the flight path, or a theoretical true AOA.

That's why it's "relative pitch angle", which is a misnomer, yes, but it just means your pitch relative to your flight path.

"This presumes that the UUA is off by a factor of two, and also that the game models this."

Game only needs to model the visuals in the cockpit. So that it shows what the real UUA would in relation to what the game itself considers AoA.

And if you know how to read the chart from the manual, you will notice, it does not actually stall until far beyond 33° indicated. Both IRL and in game.

But refer to what I said initially, at the end of the post. Stall is not the only reason why aircraft fall out of the sky. The ONE thing that could be considered wrong is that the wing rock occurs too soon at low speed.

null

[Koty]

7 hours ago, KenobiOrder said:

 

Right except what Frederf said sort of made zero sense. The chart shows a Cl increase from 1-1.4 depending on Mach number.

This makes perfect sense.

7 hours ago, KenobiOrder said:

And the claims that there would be no CL increase

Yes, in regards to how the UUA-1 compares to real AoA. Regardless the relationship, CL stays the same, all that changes is when it occurs/at what AoA. So you'd stall at 15°, 20° or 30°, but always with the same CL. Essentially you'll be stretching the CL chart horizontally, not vertically.

7 hours ago, KenobiOrder said:

also seems to assume the 28 degree line and not the 33 degree line, which shows a larger and sooner increase, and is incidentally the number we car about because 33 degrees is where the chart says wing rock starts. 

No, chart does not say wings start to rock at 33°.

7 hours ago, KenobiOrder said:

In any case its not just the Cl increase that is important here. Its the ability to maneuver generally. The current jet hits a wall at about half the AoA it should, and this means flying it is a very bad experience where you cannot pull AoA as needed to fight. 

The current jet hits a wall around where it should at normal speeds, only problem is with low speeds. The slower you go, the more AoA you should be able to pull without wing rock.

[rossmum]

9 hours ago, KenobiOrder said:

In any case its not just the Cl increase that is important here. Its the ability to maneuver generally. The current jet hits a wall at about half the AoA it should, and this means flying it is a very bad experience where you cannot pull AoA as needed to fight. 

I'm not sure what you're expecting of the jet, here. Its nose will follow anything an F-5 can do right up until the departure (which is typically happening because the wing is shadowing the tail at a high enough AoA - like Koty said, just because the wing isn't fully stalled, it doesn't mean the jet doesn't have other reasons to depart controlled flight) and the F-5 can only avoid a similar fate because it has specific design features meant to impart stability at high angles of attack. In most aircraft you'd drop a wing or spin, in the 21 you get the tail wag and maybe a rollover and tumble if you monkey the stick hard enough.

I dunno dude, plenty of us seem to have few to no issues fighting in it and it's entirely possible to outfight less experienced Hornets, despite the latter's clear (and realistic) superiority in the AoA department. Being able to fly one direction while pointing your nose isn't everything, and it's not even very good BFM, honestly.

I would also again like to draw attention that you tried to assert that the official documentation says it stalls at 28 and 33. First of all, no - as addressed by Koty. Second of all - I'm going to assume you can't read Cyrillic, but for anyone who can, it plain as day says "UUA-1" alongside each angle, aside from two marked "SUA-1" (the blinker-light warning system). Again, all values in the manual are given in the same measures the aircraft indicates. It is neither here nor there to the pilot what angle his plane is actually pointing, because all the charts are worked out for what the UUA shows, as it is the aircraft's only method of measuring angle of attack and thus has to be relied on regardless whether you find its relationship to true AoA (which as FrederF pointed out, is not strictly double) objectionable or not. If you don't understand what you're reading in the manuals, and you're willing to ignore the effects of that AoA on somewhat important things like airflow over other surfaces of the aircraft, then I don't know what to tell you because it looks like it'll just get thrown back anyway.

[KenobiOrder]

2 hours ago, Koty said:

Game only needs to model the visuals in the cockpit. So that it shows what the real UUA would in relation to what the game itself considers AoA.

This doesn't matter at all if that relationship is wrong. Which it is. The actual AoA of the wing is half what the UUA is showing. 

Let me put this very simply:

Delta Wings of the Mig-21 type do not stall at 15 degrees. You can talk all day about what you think the developer meant by local AoA. It doesnt matter. The thing that matters is that the UUA's relationship with the real AoA is wrong, because it cannot be right if it is reaching the 33 degree position in tandem with the wing hitting the true AoA at 15 degrees. 

 

[æck]

1 hour ago, KenobiOrder said:

The thing that matters is that the UUA's relationship with the real AoA is wrong

It is, even in the real aircraft. As has been mentioned several times in this thread it's due to the aircraft disturbing the direction of the airflow (remember the UUA is on the side of the fuselage). The real charts were plotted with this discrepancy included, so it doesn't matter what the true AoA is as long as the AoA as indicated on UUA matches the real charts.

Edited May 23, 2022 by æck

[KenobiOrder]

Just now, æck said:

It is, even in the real aircraft. As has been mentioned several times in this thread it's due to the aircraft disturbing the direction of the airflow (remember the UUA is on the side of the fuselage). The real charts were plotted with this discrepancy taken into account, so it doesn't matter what the true AoA is as long as the AoA as indicated on UUA matches the real charts.

It is a wild assumption that the discrepancy between the indicator and the true angle of attack is a factor of two. And it flies in the face of what is known about delta wings of the type fitted to the mig.

And that is what it matters. Because as far as the flight model acccuracy is concerned, it does not matter at what reading the indicator is at when the jet stalls.  What matters is what true angle of attack the indicator corresponds to when it reads "33 degrees"

[Koty]

4 hours ago, KenobiOrder said:

This doesn't matter at all if that relationship is wrong. Which it is. The actual AoA of the wing is half what the UUA is showing. 

Let me put this very simply:

Delta Wings of the Mig-21 type do not stall at 15 degrees. You can talk all day about what you think the developer meant by local AoA. It doesnt matter. The thing that matters is that the UUA's relationship with the real AoA is wrong, because it cannot be right if it is reaching the 33 degree position in tandem with the wing hitting the true AoA at 15 degrees. 

 

Let me put this very simply:

The 21 in DCS does not stall at 15 degrees AoA. 

[nighthawk2174]

1 hour ago, m4ti140 said:

It is a perfectly reasonable assumption, and the reason why in F-14 for instance they didn't bother and the AoA indicator is scaled in arbitrary units.

Here's a NASA paper where they were evaluating a calibration method for AoA sensors. The AoA vane was installed on a boom, one chord length in front of the wing tip. The aircraft is Piper Saratoga, whose maximum speed is close to MiG-21s stall speed, so we don't even need to worry about mach number in this case. On page 10, Table 2, you get the final conversion formulas they obtained with various methods. The measurement errors were above 30%, with zero completely off. And if you look at the raw results, you can see the relationship isn't exactly linear.

https://ntrs.nasa.gov/api/citations/20140002449/downloads/20140002449.pdf

And for the last time - MiG-21 is 50s design. Why do you expect it to get even close to angles of attack modern deltas don't let you fly at?

I don't think this paper is applicable here

1) the aoa sensor is on the wing in this paper its not on the MIG-21, additionally its on the wingtip.  As far as I understand it the effects of upwash are reduced away from the wingtip vorticies.

2) the speeds here are far less then in a mig-21 sub 100kts for half the test.

3) the wing is much thicker then the MIG-21's increasing the effects of the discrepancy.

4) The difference was minor up to 9deg aoa, which is very close to stall conditions.  Some of the test used here in DCS were at 5deg aoa with the sensor reading 10deg aoa...

5) How do we know that the aoa sensor wasn't calibrated for airframe effects?  We know that the sensor being used was only installed much latter and initially only the probe aoa sensor was used.

6)The difference in game is always 50%, in the paper it grows from nothing to 25% max at/near stall.  In game if it were a 25% difference that'd be a sensor reading of 33deg for a true aoa of ~25deg.  Which is much higher then it is now.

Edited May 23, 2022 by nighthawk2174

[Koty]

Difference in the paper is smaller, because its not directly on a thick, cylindrical fuselage, but in relatively free air.

How do we know it wasn't calibrated?

Here is the error equation for 23ML and 23UB

The way to read this, on the left you have indicated AoA, alpha-phi is true AoA.

So on the ML the the indicated (or local) AoA shows double the true AoA - 5.5°, on the UB with slightly skinnier and pointier nose it gives you 1.6x true AoA -1°. The manual makes a fairly big point of the AoA being the local AoA in the area of nose, and with no correction. And again, its on the fuselage NOT THE WING. So comparing wing thickness to the 21's wing thickness is pointless, compare it to the fuselage the sensor is on, which is, I would say, considerably thicker than wings.

 

You could however figure that out with common sense. Why would the manual advise 12° AoA at landing, when 10° pitch is tailstrike territory?  

Edited May 23, 2022 by Koty

[Frederf]

There is a NACA paper about correction values of local AOA of fuselage mounted sensor (and other locations). It is not negligible. There is a reason they use super long temporary booms on flight aero testing to get as close to free stream data as possible.

Btw DUA signal is transmitted electroniclly. It would be easy for a multiplier circuit to adjust the scaling of the value of the sensor electrically. When you are grabbing gauges out of the Soviet parts bin you might desire an expanded display without fractional values.

[KenobiOrder]

17 hours ago, Koty said:

Let me put this very simply:

The 21 in DCS does not stall at 15 degrees AoA. 

Lol yes it does. It stalls in game at a true aoa of abiut 15-16.

[KenobiOrder]

15 hours ago, m4ti140 said:

1. And the effect from a massive barrel going through the air is non existent?

2. I don't see how higher speeds make it better

3. The wing is not the problem with the 21 sensor

4. It's not. The error becomes 0 around 4 degrees, it's higher both below and above that value. Also the errors for a vane mounted the side of the forward fuselage will be completely different than in wing upwash region.

5. Follow the exact take-off procedure in the real flight manual and use info bar instead of UUA. Your tailpipe will hit the ground.

6. For the last time, it's a different mounting point. It was meant to illustrate the fact that the flow around the aircraft is completely different than free stream.

 

Except the issue at hand isn't whether there is an error. It's how big the error is. The current in game error is greater than twice the true aoa. None of the evidence provided suggests that this is accurate. Both the paper you linked and the comparison to the mig23 yield much lower discrepancies than we have in game. The mig23 isn't the same plane, but continuing your comparison, that error equation only suggests that the amount of error in the sim is excessive. Using the indicated stall of 33 degrees, it would give a true aoa of between 19-and 21.25. Which would be a much more reasonable stall aoa for a delta swept at 57 degrees.

[Ala13_ManOWar]

It's worthless to keep trying to explain things mate, he doesn't understand, he doesn't want to understand. The precise points have been presented to him, it's up to the flatearther trying to understand those points and clear data or just go full denial and not seeing or even try at all just because.

At some point it's better for one's health just let it go and stop trying to deal with these kind of individuals all the time by own experience.

[Skuva]

Upon reading a good chunk of the flight operating instructions it realy seems that the UUA-1 gives a very drastic deviation, at least for the low speed scenarios mentioned.

I encountered the same lift coefficient graph Frederf posted (but translated). And that got me thinking, isn't this basically supporting KenobiOrder's argument? at least for low speeds. It suggests that at mach 0.5 the stall occurs at well above the indicated 33° (I would guess above 40°, or 20° on the convertion to "real" AoA). Is there another effect that would cause the rocking in this situation before reaching such point?

 

Edited May 25, 2022 by Skuva

[Frederf]

"Stall" means maximum CL. That's the definition. It is independent of controllability or instrument value. The argument put forth was that the freestream AOA and the UUA-1 indication should be or more closely be 1:1 i.e. 16 freestream AOA = 16 on the gauge. Clearly in the landing regime that's impossible because the flight manual describes UUA-11 readings of 15 or so. The idea of 1:1 degrees:degrees is interesting but I haven't seen any evidence for that.

The other notion is that a stall landing would be at essentially 28-32° body angle which can't be true either. Back in the day the AI flew "UUA-1" AOA according to the F2 value and it was silly. They were at double-triple-quadruple body angle in level cruise compared to pilot. Notice that the flight manual says that indication can depend if on 24V battery or 28V ground/engine power by about that ratio.

At 360 km/h approach speed (0.3M), Cy max is about 1.2 assuming it can even be reached. What CL is needed for approach path? Assuming 73.5kN of lift, 360km/h 23m2, 1.225kg/m3 it's about 0.52. Assuming (again) the UUA lines are linear spacing I would put the UUA-1 reading at 18°. Is that what we get?

[Koty]

See, here is one thing that has been bothering me.
IF the relation is wrong, then what specifically is the correct relation. And more importantly, according to what?

[KenobiOrder]

On 5/24/2022 at 1:10 PM, m4ti140 said:

What do you mean "lower discrepancy"? The discrepancy is by a factor of 2, just like in MiG-21bis! The zero point is just moved. It's not the same aircraft, the way it's mounted can be different even. And at this discrepancy you would still already be dangerously close to a tailstrike if you followed the real takeoff procedure. Please show me a piece of documentation showing the real critical AoA is higher.

The flight characteristics book for the 21 refers to loss of conventional lift as a stall, because aircraft would tumble at this point, vortex lift regime is not even mentioned there. I don't know why the aircraft becomes uncontrollable by this point (and couldn't find anything in documentation), but it does.

In DCS it doesn't actually even stall at all at this point, if you can keep wings level and pitch further keep flying like that. In fact you can keep flying way past reasonable angles of attack into region when the aircraft would surely tumble due to vortex shedding.

Except its not by a factor of two. Not sure why you bothered pointing out that they are different planes, since I already stated that. But since you want to use the Mig-23 error equations as an example, I am simply continuing that example. And they do not show the AoA being off by a factor of two. Those error equations yield errors of about 40%. If you plug them into the Mig-21, you get stall angles of attack of around 20 degrees. 

And no, the planes stalls when it hits 33 degrees in game, except at very very very low speeds. The stall wing rock is clearly some kind of scripted behavior. It also doesn't make sense that the critical AoA changes based on speed, because that is not how wings work. 

[KenobiOrder]

15 hours ago, Frederf said:

"Stall" means maximum CL. That's the definition. It is independent of controllability or instrument value. The argument put forth was that the freestream AOA and the UUA-1 indication should be or more closely be 1:1 i.e. 16 freestream AOA = 16 on the gauge. Clearly in the landing regime that's impossible because the flight manual describes UUA-11 readings of 15 or so. The idea of 1:1 degrees:degrees is interesting but I haven't seen any evidence for that.

The other notion is that a stall landing would be at essentially 28-32° body angle which can't be true either. Back in the day the AI flew "UUA-1" AOA according to the F2 value and it was silly. They were at double-triple-quadruple body angle in level cruise compared to pilot. Notice that the flight manual says that indication can depend if on 24V battery or 28V ground/engine power by about that ratio.

At 360 km/h approach speed (0.3M), Cy max is about 1.2 assuming it can even be reached. What CL is needed for approach path? Assuming 73.5kN of lift, 360km/h 23m2, 1.225kg/m3 it's about 0.52. Assuming (again) the UUA lines are linear spacing I would put the UUA-1 reading at 18°. Is that what we get?

The original main point was simply that the explanation given by the developer was not correct, as F2 view does not show pitch angle. It shows true AoA. 

The reason that the stalling AoA of 15-16 is wrong is because a delta of the Mig's type does not stall anywhere close to this. It simply does not. 

The issue with the landing AoA being off assumes either 1:1 or that the error relationship should be linear. The error of the wing mounted gauge in the test was not linear. Landing angles of attack vs read out on the UUA could be much smaller. Although it isnt obvious to me why one would land at critical angle of attack. 

In fact what might be wrong here is that the relationship in the game does in fact seem linear, as it appears to be two to one at all angles of attack. 

Edited May 26, 2022 by KenobiOrder

[Frederf]

Eh? That's Tsagi S-12 alpha sweep.

UUA-1 indication is according to this formula in DCS: (Y = 1.96X + 1.41) It's not "double". I agree that this is almost certainly a weak simulation of the actual instrument. The relationship between UUA-1 indication and free stream AOA is certainly not this simple linear equation at all Mach. However, put a Post-It over the gauge and check the airplane performance for reasonableness. It looks much better in these cases.

[KenobiOrder]

45 minutes ago, Frederf said:

Eh? That's Tsagi S-12 alpha sweep.

UUA-1 indication is according to this formula in DCS: (Y = 1.96X + 1.41) It's not "double". I agree that this is almost certainly a weak simulation of the actual instrument. The relationship between UUA-1 indication and free stream AOA is certainly not this simple linear equation at all Mach. However, put a Post-It over the gauge and check the airplane performance for reasonableness. It looks much better in these cases.

Yes that's for the airfoil, not the entire wing. If you reference the chart from the U.S manual it shows the actual stall at a much higher aoa than 33 indicated. There is no mention of wing rock except at speeds at 250kph and below, making me wonder if the dcs mig 21 is modeled backwards by accident.

[Frederf]

Have Donut? MiG-21US PFOI? Which diagram number? "Wing rock" is not a stall condition. That's controllability or stability.

[KenobiOrder]

15 hours ago, Frederf said:

Have Donut? MiG-21US PFOI? Which diagram number? "Wing rock" is not a stall condition. That's controllability or stability.

Wing rock is a stall condition. It generally occurs because one wing stalls first, then the other, and then oscillates. In any case the US manual mentions no wing rock whatsoever until you get below something like 250kph, at which point it reports +-50 degree wing rock. The only pre-stall condition reported is buffet, which itself only occurs in certain conditions. As can be clearly see in the stall diagram, on both Russian and American documents, the only warning is for buffet. There is no wing rock warning on the chart, and no mention of it at higher speeds in the expanded description that I could find.  What can also be seen in that the 28 and 33 degree limits are NOT the stall limits, unlike in DCS Mig-21. The instructions call the 33 degee limit the stall limit, but thats contradictory to the chart....except at high mach numbers. I sort of get the impression that the 33 degree limit is placarded as a safety limit because the stall AoA appears to decrease substantially with mach number.  

nullnullnullnullnullnull

[Frederf]

You don't have to post the manual at me. I've read them dozens since the module was first released and experienced every public flight model change over that time.

What is your assertion? There are three (to keep it simple) variables here: A. UUA-1 B. freestream "F2" AOA C. Cy.

All the charts in the world showing BC relationship tells nothing about AB relationship. The only factoid I can think of is that takeoff pitch of 5° gives some 11-13 UUA-1.

Which relationship AB BC or AC do think is wrong. I definitely think something isn't right since AB is Mach independent in DCS but I want you to draw those graphs how you think this airplane is so we can evaluate them.