5 hours ago, kcmo said:

Thanks for all your help, guys!

I put on my test pilot hat this morning and ran the test again on the f-18. Twice I couldn't get it to recover.

Then I decided to change my primary "recovery indicator" from the airspeed indicator to the AOA indicator. 

I am humbled to see that, when trying to recover, you can reach a flyable airspeed and yet STILL be flying at too high an AOA.

I thought I was reaching a reasonable attitude to start pulling in back pressure, but I had been ignoring the fact I could be at full power with nose down, and STILL have to wait until the relative wind returns to a laminar flow over the wings.

Thanks again for walking me through this...

If you're really curious, have you tried DCS?  I think it gets dismissed too easily by some in the civilian sim community as a "shoot em up" game - I know I was guilty of this for a couple years.  But it actually has a stellar flight model; I've not flown an F/A-18, but the DCS Hornet has the best-feeling flight model of any jet I have ever tried, in ANY home sim.  I mean it feels like a 30 - something thousand pound jet.  It is fantastic; FAR better than the MSFS 2020 Hornet, I mean night and day different.  Given the intended use of the software, DCS makes a real effort at edge-of-envelope departure modeling, energy bleed under g, and proper FCS behavior to help with all that. 

HIGHLY recommend the DCS Hornet (or any other DCS jet) if you're curious about all this.  The base sim is free to install, and they run two week free trials on any aircraft, so you can give it a two week try for absolutely free (just install direct from Eagle Dynamics, not Steam, as I think the trials don't work with the Steam version?  Someone can correct me there though.)

56 minutes ago, Stearmandriver said:

Have you tried DCS?

Yeah, I own it but the flight model always felt brutally unforgiving - which is to say harder than reality. Thanks for bringing it up though - I'll check it out again!

 

 

4 minutes ago, kcmo said:

Yeah, I own it but the flight model always felt brutally unforgiving - which is to say harder than reality. Thanks for bringing it up though - I'll check it out again!

 

 

It's hard in a sim to realize how hard you're pulling because we can't feel the g.  It's a much more visceral thing in reality - you can absolutely feel how hard the wing is clawing the air when you're being squashed haha.  I'm not a pointy nosed jet guy in reality but I have taught aerobatics for years so I have some feel for energy bleed and accelerated departures, but I still do it to myself in DCS often enough 😉.  If I keep an eye on g as I'm doing it though, it seems plenty realistic.  It's just a maneuverable airplane and we can't feel it.

On 12/24/2024 at 7:43 PM, Bobsk8 said:

When I was studying for my PPL, may years ago, I got a book called Stick and Rudder by Langewiesche which I read cover to cover several times. I always remember that he stated that  you could put a piece of tape on the yoke post just prior to the aircraft stalling , and if you never went past that tape when pulling on the yoke, you would never stall the aircraft no matter what your airspeed was. I used to think about that everytime I would slow fly. 

Getting back to this part of the discussion.

I don't have "Stick and Rudder", but I think I finally need to order it.

In the meantime, I do think the description above is accurate (when CG is fixed), and I can explain why.

Here's an argument why, for a fixed CG, a given elevator position corresponds to the same angle of attack, independent of load factor. In particular, this is then also true for the critical angle of attack, meaning that the aircraft stalls at the same elevator position, independent of load factor.

Let's say we're flying along at a load factor of 1. The wing is producing a given amount of lift L_w, and we're holding the elevator in a suitable position so that the horizontal stabilizer is producing a downforce L_s that results in a zero pitching moment overall. (We use the convention that L_s is negative because the lift from the horizontal stabilizer acts downwards.) In other words,

r_w * L_w + r_s * L_s = 0

where r_w is the moment arm from the CG to the wing's center of lift, and r_s is the moment arm from the CG to the horizontal stabilizer's center of lift.

Now let's say that, at the same angle of attack, we double the airspeed, resulting in a load factor of 4. (For example, say we're in a 75-degree angle-of-bank turn. I'm using these particular values because they result in round numbers.) Lift increases with the square of the airspeed, so both the lift from the wing and the downforce from the horizontal stabilizer increase by a factor of 4. Because both increase by the same factor, the pitching moments still balance out:

r_w * 4 * L_w + r_s * 4 * L_s = 4 * (r_w * L_w + r_s * L_s) = 0

This means that elevator position remains the same when flying at this higher load factor with the same angle of attack¹.

(Of course, in practice, while we're transitioning from one state to the other, we would be making significant manipulations of the flight controls, and we would need to increase engine power considerably to maintain the doubled airspeed. What we're interested at here is the state after we've achieved equilibrium at the increased airspeed and load factor. Alternatively, we can imagine we're in a wind tunnel, where it's possible to simply increase airspeed while keeping the other factors -- angle of attack and elevator position -- constant.)

Generalizing, we can say that a given angle of attack goes with a corresponding elevator position, independent of load factor. This is true specifically for the critical angle of attack, meaning that the stall occurs at the same elevator position, independent of load factor.

All of the above assumes that CG is kept fixed. When CG changes, the moment arms r_w and r_s change, and hence the elevator position that corresponds to a given angle of attack also changes.

Of course, as @Stearmandriver points out, the best tool for understanding angle of attack and the stall is an actual angle of attack indicator.

--------

¹ Wait. When we fly steep turns, don't we need to exert a significant pull to maintain altitude? Absolutely. This is because steep turns are usually flown at the same airspeed that we previously had in straight-and-level flight, so to produce the additional lift required in the turn, we need to increase the angle of attack, and hence pull. In contrast, in our discussion here, we produce the additional lift by flying the turn at a much higher airspeed than we previously had in straight-and-level flight, which in return allows us to use the same angle of attack (and, as we've seen, the same elevator position) that we previously had in straight-and-level flight at the lower airspeed.

Edited December 28, 20241 yr by martinboehme

Interesting discussion!  I *think*, to the extent that I can follow the math, that I can agree with you in a very specific circumstance of a given aircraft (not a type) at a given cg, if this plane does not have a trimmable stab but uses elevator trim tabs.

I was preparing to disagree about load factor and point out the importance of transient states vs steady states.  I was thinking of the example of diving for loop entry speed in, say, a decathlon, and giving the stick a good pop for an initial 4g pull for a nice crisp entry; and thinking that if I was in slow flight and made the same stick deflection that would be a certain departure.  But I think that in this example, I would be starting with the stick further back in slow flight (because of trim state) and so a similar stick deflection would bring the stick further aft than in the loop entry.  So I can see how this is theoretically possible.

I think also though, it needs to be recognized that this is simply an academic curiosity and nothing that could be used in reality, because in reality, cg is hardly ever the same from flight to flight and continuously changes in flight as fuel is burned - to a varying extent in different aircraft, but it is always changing.   Thus, even if you only ever flew one tail number, that did not have a trimmable stab, and you always flew solo, this STILL would not be a useful metric for stall avoidance - you'd always have a different fuel state and your aircraft load would always be a little different, even if it's just the winter coat you threw in the back with the tow bar and chocks that weren't back there last time, etc. 

I have to assume that if this is mentioned in stick and rudder, it's in there specifically as an academic example of how an aircraft will always stall at the critical AoA vs airspeed, and he stresses it as such.  From a safety culture standpoint, there are a lot of reasons we wouldn't want people actually trying to use the metric of stick position for stall avoidance.  There are better parameters.

Thanks for doing the legwork on formulizing this though; an interesting exercise.

2 hours ago, Stearmandriver said:

Interesting discussion!  I *think*, to the extent that I can follow the math, that I can agree with you in a very specific circumstance of a given aircraft (not a type) at a given cg, if this plane does not have a trimmable stab but uses elevator trim tabs.

Agreed -- in particular that a trimmable stabilizer invalidates the whole argument because it allows the downforce generated by the stabilizer to be changed without changing elevator position.

2 hours ago, Stearmandriver said:

I was preparing to disagree about load factor and point out the importance of transient states vs steady states.  I was thinking of the example of diving for loop entry speed in, say, a decathlon, and giving the stick a good pop for an initial 4g pull for a nice crisp entry; and thinking that if I was in slow flight and made the same stick deflection that would be a certain departure.  But I think that in this example, I would be starting with the stick further back in slow flight (because of trim state) and so a similar stick deflection would bring the stick further aft than in the loop entry.

Yes, I think this is accurate.

Also agree that the argument only applies in the steady state. I believe that after a change in elevator position, a steady state with respect to AoA is reached relatively quickly though -- my intuition is that this would probably take just a few tenths of a second, certainly much faster than reaching a steady state with respect to airspeed or attitude. But I'd be curious to know what the actual numbers are.

2 hours ago, Stearmandriver said:

I think also though, it needs to be recognized that this is simply an academic curiosity and nothing that could be used in reality, because in reality, cg is hardly ever the same from flight to flight and continuously changes in flight as fuel is burned - to a varying extent in different aircraft, but it is always changing.   Thus, even if you only ever flew one tail number, that did not have a trimmable stab, and you always flew solo, this STILL would not be a useful metric for stall avoidance - you'd always have a different fuel state and your aircraft load would always be a little different, even if it's just the winter coat you threw in the back with the tow bar and chocks that weren't back there last time, etc.

Agreed. I think another big factor is that pilots tend to be much more aware of stick pressure than stick position -- and stick pressure for the same AoA varies greatly of course with trim and airspeed.

2 hours ago, Stearmandriver said:

I have to assume that if this is mentioned in stick and rudder, it's in there specifically as an academic example of how an aircraft will always stall at the critical AoA vs airspeed, and he stresses it as such.  From a safety culture standpoint, there are a lot of reasons we wouldn't want people actually trying to use the metric of stick position for stall avoidance.  There are better parameters.

I'm curious to look for this discussion in Stick and Rudder and, if I find it, see how it is presented. I share your expectation that it would be presented as an aid to understanding rather than as a practical stall avoidance technique.

Thank you for the discussion!

1 hour ago, martinboehme said:

 

I'm curious to look for this discussion in Stick and Rudder and, if I find it, see how it is presented. I share your expectation that it would be presented as an aid to understanding rather than as a practical stall avoidance technique.

Thank you for the discussion!

All I can remember, reading this 45 or more years ago, is  that a danger signal is when you pull the yoke towards you too far. At some point, the aircraft will stall. I once flew with a newly licensed pilot, and he overshot the final in the pattern, tried to turn back, and keep his altitude by " pulling the yoke towards his stomach" (trying to maintain altitude as the bank angle increased)  I could see what was about to happen, and I reached up and pushed the yoke forwards just as the stall horn started to bleep. It was a classic case of stall, spin entry in the pattern which would have killed us both. I remember thinking at the time, that Stick and Rudder , saved my butt. 

4 hours ago, martinboehme said:

Agreed. I think another big factor is that pilots tend to be much more aware of stick pressure than stick position -- and stick pressure for the same AoA varies greatly of course with trim and airspeed.

I think so too - stick pressure, and the amount of stick deflection from its neutral point, wherever that is trimmed to be.  That's why it initially seemed to me that using an equal "yank" that I would use to enter a loop at 160mph would stall me in slow flight - I was thinking of it as the amount of stick displacement.  Maybe this is just me personally, but I don't think I have much awareness of the stick neutral point changing with trim state.  I know that it does (again, in a plane using trim tabs) but I just don't consciously take note of that.  Flying to me is more about pressures and deflection from there / feeling the g, and how these interact with whatever energy state I'm in at the time.

From a practical standpoint, a pilot will be better served by becoming familiar with how their airplane feels during normal and accelerated stalls.  When you're familiar with the mushiness, the change in control effectiveness, and maybe the buffet as you approach a stall (most planes buffet in an accelerated stall, that "wrapped up" overshooting turn to final), unloading that wing becomes instinctive.

Just saw this, and it seemed relevant:

https://www.avweb.com/aviation-news/faa-recommends-adding-aoa-to-all-airplanes/

A few days ago, the FAA published a Special Airworthiness Information Bulletin that explicitly recommends retrofitting AoA indicators to all Part 23 airplanes.

On 12/24/2024 at 7:43 PM, Bobsk8 said:

When I was studying for my PPL, may years ago, I got a book called Stick and Rudder by Langewiesche which I read cover to cover several times. I always remember that he stated that  you could put a piece of tape on the yoke post just prior to the aircraft stalling , and if you never went past that tape when pulling on the yoke, you would never stall the aircraft no matter what your airspeed was. I used to think about that everytime I would slow fly. 

I now have my copy of Stick and Rudder, and I've found the section that you referenced. I found it interesting reading -- it touches most of the points we've already discussed here, plus at least one that we haven't, so I thought I'd give a summary.

The section in question is called "Stick Position" and is in chapter 4, "The Flying Instinct". In my edition, it starts on page 65.

First of all, here are some excerpts on how Langewiesche discusses the relationship between stick position and AoA:

"The most correct way, in the engineering sense, of gauging one's buoyancy is by the position, rather than the feel, of the stick. [...] Thus stick position, since it causes Angle of Attack, also indicates Angle of Attack, and thus indicates what the pilot wants to know -- his buoyancy or 'lift', that is, how far he is from the stall."

But he acknowledges that it is hard to gauge stick position in practice:

"There is a catch, however, to stick position as an indicator of buoyancy. [...] He [the pilot] is much more conscious of stick pressure than of stick position. Unless he actually looks at his hands, he may bring them quite close to his stomach and not realize it [...]."

He therefore proposes that trainers should have an indicator showing the position of the elevators, which would serve as a proxy for AoA:

"A training ship should have an indicator that would show, by pointer and dial on the instrument board, in what position the pilot is holding the stick [...]. Such an indicator could then be calibrated in terms of Angle of Attack, and also in terms of buoyancy and closeness to the stall."

At the time, I assume it was harder to construct an actual angle of attack indicator; if Langewiesche had had access to a modern AoA indicator, I assume he would simply have advocated for that -- so I think his ideas are very much along of the lines of what @Stearmandriver argues for.

Langewiesche also makes it clear that elevator position does not agree perfectly with AoA:

"Another catch to using stick position as an indicator of buoyancy is this: our preesent airplanes are not entirely well behaved. Their Angle of Attack is controlled, true enough, by stick position; but it is haphazardly influenced also by throttle setting and (to a lesser degree) by the loading condition of the airplane."

We had already discussed the influence of CG, but I hadn't thought about the influence that throttle setting has. He expands on this:

"Most of our civilian airplanes will not stall with power off unless the pilot holds the stick clear back against his stomach; with power on many of them will stall when the stick is held only halfway back. This is because the propeller blast hits the upward-deflected flippers and makes them unduly powerful; and in some airplanes because the propeller thrust, pulling forward low on the airplane coupled with the drag, pulling backward higher up on the airplane, tends to force the airplane to higher Angle of Attack [...]."

Yes, they used to call the elevators flippers; how I wish we'd kept that term. The term "buoyancy" also seems to have been more common at the time than "lift". Here I prefer the modern term, because it's shorter and, unlike "buoyancy", doesn't invoke an idea of "lighter than air" flight.

I've only read this small bit of the book so far, but I can see why it's so highly regarded, and I'm looking forward to reading the rest.