Going Beyond Procedure — an open letter
In response to Juan Browne’s concern for spins in light twin airplanes as shown in the BlancOlirio(1) YouTube channel
In the video covering the Beechcraft Travel Air crash of N369BB(2), Juan Browne addresses concerns for asymmetric thrust in light twin aircraft with specific focus to the minimum controllable velocity (Vmc), spin awareness, instructor awareness for spin avoidance, and spin recovery. He emphasizes these with a video for the Beechcraft Baron crash of N7345R(3). In these videos I think he makes some really interesting points including points to emphasize:
- instructors should make a rudder block for Vmc Demo exercises via a foot inhibiting full range of pedal motion therefore limiting rudder deflection hence increasing “Vmc”
- why the light twin susceptibility to spin during Vmc Demo with normally aspirating engines losing power with increased density altitude hence Vmc decreasing through stall speed leading to stalls with high yaw potential.
- the importance of center of gravity with more forward cg being advantageous to recovery
- the vertical stabilizer and rudder could stall
I also think he presents some significant errors regarding spin understanding that should be addressed. I believe these result from oversimplification that has been ingrained in the aviation community. I intend to address these. Aileron has merit in spin recovery as does asymmetric thrust, but these can be confusing hence the FAA choice not to teach them hence the aviation community discounting them.
Having said this, I really like Juan’s point and am re-emphasizing again the instructor technique of creating an artificial “full rudder” with a faster than actual Vmc by putting a foot to create a less than actual rudder capacity training rudder limit. DPEs should probably do this too. Staggered feet so as to also block wrong rudder direction may also be considered. In addition to this and if no angle of attack (AOA) display available, I’d recommend doing your power off stall at the intended Vmc Demo altitude just prior to Vmc Demo so as to know what current condition stall speed is; add ten knots to this as a go no slower than speed to use as a hard limit honoring even if not fully getting through the Vmc Demo. Be sure to accelerate well above both this and published Vmc before starting the Vmc Demo. Better, if you have AOA displayed, use AOA, don’t slow through the cautionary range even if not yet full rudder and 5 degrees AOB into “good” engine. Juan is quite correct, avoid nearing stall as you’re getting to high yaw while stall plus yaw equals spin.
As Juan said, how did we get to reducing Vmc to and to less than stall? I’m really glad he reviewed this topic as I myself had not previously thought of such. Below I have appended a screen shot of Juan’s screen shot from the Baron flight manual highlighting the issue. While looking at this, consider Sammy Mason(4) as he wrote,
Inadvertent spins can occur during the Vmc phase of twin-engine operation. In normally aspirated engines, horsepower diminishes with increased altitude. This results in a weakened yaw reaction from asymmetric power. This means that the yaw can be controlled at a lower indicated airspeed. As altitude increases, the minimum control speed comes closer to the stall speed until, finally, the two meet. A spin is almost a certainty when a pilot loses directional control while entering a stall. Also, at these higher altitudes, the forces that induced autorotation produce a more violent spin entry in the less dense atmosphere.
While Juan mentioned difficulty recovering from spins due to aft cg as likely in the Baron crash, he discussed the moment arm of the rudder getting reduced, Mason adds voice to further aft cg concerns on the same page in which he addressed Juan’s concern for increasing altitude decreasing Vmc with,
An extremely aft cg can increase the Vmc speed. The distance between the cg and the rudder is reduced, decreasing the effective moment arm and reducing its effectiveness. Spins are more easily excited at aft cg because it is easier to obtain a stalling angle of attack. Also, as a rule, flat spins are more easily induced at the aft cg.
Though not specific to spin avoidance nor to spin recovery, I also like Juan’s hypothesis toward why a perceived(5) increase in light twin accidents. What he mentioned with his suspicion is an example of the “Cobra Effect”(6) having occurred with airline transport pilot (ATP) licensing rules changes following the Buffalo crash(7) in which the FAA increased hours requirements for pilots. Due to these, Juan believes much more Vmc training has been shifted from simulator environments into aircraft while a pilot demand has rapidly amplified the numbers of these events while due to an upward flow of personnel from learning to instructing to leaving for airline jobs, the safety knowledge has diminished. We are training with riskier methods while being trained by more ignorant instructors. Policy decisions will always carry unexpected unintended consequences(8) while sometimes these reactions will be completely opposite intended effects. Wanting to reduce snake bite fatalities, governing bodies in certain Indian cities offered bounty for cobra skins. Rather than encouraging hunting, however, the immediate effect was to farm snakes thus increasing the total population in the area and increasing risk exposure. Cobra Effect.
All in all, Juan’s discussion as to why the spins and spin avoidance is outstanding. His efforts at spin recovery, however, are lacking.
Consider first the Travel Air crash in which we hear in the video the crash aircrew calling an emergency to air traffic control (ATC). In this, Juan presumes the crew was lost with no idea what to do and called first thus violating airmen priorities to Aviate, then Navigate, and lastly Communicate. Yet the procedures for spin recovery are rather short and can be executed rapidly. Juan made a presumption. We don’t know if procedures were executed nor if anything else were attempted after such may have proved inadequate. The crew could have already executed the procedure while not seeing any improvement and thus be stuck for ideas as to what to do. A call could have been subsequent not initial. Just holding the procedure and hoping for improvement seems to me to be a particularly active version of one of the FAA’s identified hazardous attitudes, this being Resignation.
Running to the end of the procedure then accepting being stuck is Active Resignation. This despite what the Baron manual may say in its procedure regarding holding input till improvement. With any spin procedure, you should notice rather quickly a shift in equilibrium else a transition to a different spin mode else an incurring of or change in oscillations. The only reason for lengthy hold would be statically stable dynamically unstable oscillations. Such would indicate opportunity for a shift in equilibrium or mode given time. Nothing else does.
Juan refers to some spin training he did in a Pitts(9). He provides us a link to a video from a similar event. In both videos concerning light twin crashing from spins, he says “adding power flattens spins” which he shows from the Pitts video. He’ll further emphasize “Never Aileron!” Here we have problems. One cannot merely compare aircraft and presume lessons from one transfer regarding out-of-control flight and spins. One must compare and contrast the two airplanes in question.
Why might Juan say “never aileron” for spin recoveries? Partly because spin training gets simplified so as to be easy to remember. Most spin training is done in light general aviation aircraft or in aerobatic aircraft. Though I disagree with such thinking, these aircraft are often considered “no load” aircraft. By no load, I mean the mass distribution between wings versus fuselage is fairly equal. I disagree with such as even for totally equal, you’re never no load. I prefer either “light loaded” or “equally loaded.” The point here, however, is that in these aircraft, aerodynamic forces dominate inertial forces. Inertial forces in these balanced loaded aircraft will still resist spin recovery but they tend to cancel each other out in terms of gyroscopic responses to other added forces hence why the aerodynamic flow alone should be viewed in their recovery. In these aircraft, generally elevator is more than up to task to break the stall while rudder often is up to task to break the yaw without outside help though blanking of one and/or the other can happen in some GA designs. Inertial forces play, but they’re minor players. Since they’re generally balanced, actions you might take based on inertial forces don’t apply. Yet you’re still susceptible to conservation of momentum and centrifugal forces. As inertial concerns are smaller players and as elevator and/or rudder are up to task in spin recovery, inertial concerns get dropped from discussion and from procedures to simplify them in order to make easier to remember and quicker to execute tools for the pilot.
The FAA gives us “PARE” for spin recovery (though I believe Rich Stowell coined the acronym):
Power — Idle
Aileron — Neutral
Rudder — Against Yaw (against Turn Needle)
Elevator — Forward (erect or upright spin (high AOA)) / Aft (inverted spin (negative AOA though most displays peg at zero))
A Travel Air and a Baron are a light twins, they have engines and fuel tanks on the wings with associated support structure. Juan correctly tells us this in both videos in the context of conservation of momentum. He forgets, however, to contrast the Pitts mass distribution. A Pitts is a biplane single engine tandem single seat or tandem seat aircraft that has its fuel tank between the aircrew and engine. The wings are short; the mass distribution of the Pitts is in the fuselage. This mass is mostly on the longitudinal axis while the Travel Air’s and Baron’s are mostly along their lateral axes. One is “fuselage loaded” while the others are “wing loaded.” Further, the Pitts is highly susceptible to the four turning tendencies of propeller aircraft.(10)
With the Travel Air, turning tendencies exist though they’re much less impact than thrust asymmetries if power positions get split. Juan correctly hits the mass distribution of the light twin but he only does this in the context of conservation of momentum in that he highlights the rudder will have difficulty overcoming rotation once that large “flywheel” gets spinning. He also misses that there are three flywheels at play in any spin.
For a moment we’ll consider aerodynamics only. We’ll discuss what ailerons do, what power does. Then we’ll look at how different inertial distributions play. From these, you will see that should PARE fail, yes aileron is valid while in the twin power may be too. Yet you’ll also see how confusing these can be and how they could lead to more trouble if done wrong. Hence, you will appreciate why the FAA doesn’t normally discuss these. Then we’ll come up with guidelines you can adapt to specific planes so as to have these considerations in mind yet not need to try to think through them realtime.
Imagine we’re in a spin. Juan said “never aileron.” Why? If you put the stick against the spin trying to undo the spin, then you’ve increased the AOA on the inside retreating more stalled wing hence driving it further into stall thus both further reducing its lift and much more drastically increasing its drag. Makes sense aerodynamically as to why after rudder against yaw or stepping opposite the turn needle, you wouldn’t necessarily want to do the same with the stick. But if stick away proves bad, why not stick into yaw aka with turn needle? The same logic tells us stick into erect spins makes the amount of stall difference between the wings less and the yaw less as less drag on the inside wing decreases as its stall reduced. This would mean in stalled conditions our lateral stick would be reverse command. United States Air Force Test Pilot School (USAF TPS) shows us there’s actually a different airflow concern going on in which we won’t have reverse command here:
What we get from an aileron deflection is not a steady state roll rate, but rather a new orientation for the aircraft on the spin axis. The effect is also different from what you might expect from a lifting surface operating beyond the stall angle of attack. When applying lateral stick in the direction of the spin, instead of the higher angle of attack aileron having less lift than the lower angle of attack aileron and causing a roll away from the spin, the aircraft tilts around the x axis towards the spin. This happens not because of lift over the aileron surface, but rather because the downward deflected aileron traps a pocket of higher pressure air than the upward deflected aileron which allows some airflow to slip through. This effect quickly neutralizes after the aircraft has banked just a few degrees. Nonetheless, the resulting bank angle now means the aircraft has a pitch rate.(11)
Despite this, the US Navy (USN) in its Out of Control (OCF) Flight Training Instruction (FTI) for the T-6 Texan II gives:
Generally speaking, ailerons are not very effective in light aircraft at stalled AOA and should not be used for entry or recovery. In fact, application of ailerons creates a yawing motion in the opposite direction, known as adverse yaw. At spin initiation, a cross-control situation enhances spin entry. Conversely, deflection of ailerons into the spin reduces the autorotation rolling moment by reducing the AOA on the “inside” wing and can produce the adverse yaw necessary to aid rudder yawing moments to affect recovery.(12)
As you can see, all of this is a bit confusing (USAF TPS is correct). It gets worse for an inverted spin as roll is opposite yaw. While inverted, if you’re yawing nose right, then the right wing will rise relative to airframe hence you’re in a left roll. Any lateral stick thoughts flip-flop as the wing is upside down hence the ailerons need to drive the other way to get earth normal wing shaping. Similarly, the elevator aft functions as did elevator forward.
Not only can you get it wrong with aileron, even getting it right could have wrong secondary effects. You can have too much of a good thing. Presuming light loaded airplanes in upright spin, Mason writes:
Ordinarily, the use of aileron with, or in the direction of, the spin will speed up the rotation but will not flatten the spin. In most cases the spin angle will be steeper. The increased rotational speed may result in a delayed recovery.(13)
One of the easiest ways to flatten a spin and make it unrecoverable in some airplanes is to use aileron opposite to spin rotation. The use of aileron in the direction of the spin (right spin, right aileron) will usually increase the speed of rotation. As a result, it may take a little longer to stop the rotation during recovery. However, because pro-spin aileron tends to develop a healthy amount of roll relative to yaw, the spin is less likely to flatten than if aileron opposite rotation is used.(14)
As you don’t need the aileron for most light loaded aircraft recovery, the common thinking reasonably suggests why risk getting it wrong. It is a valid point. It does lead us to discount the physical reality of ailerons however. Were you to not be recovering after executing PARE, you might want to consider them. We’ll get a better sense of how to use them when we consider all factors at play though I like the thought here that if you have nothing left to try, what do you have to lose for trying. If you’re stuck, do something, if that makes it worse, do the opposite.
With this, Juan has proven correct in his adamant stance for no aileron only in context. I counter both with what to do after PARE has proven unsuccessful and with caution to standby till we look more at mass distribution. Note that most our spin discussions are for doing intentional spins with lightly loaded or equally loaded airplanes that tend to readily spin and therefore tend to recover easily. Yet our concern is for aircraft that are less cooperative in recovery to which most discussion is lacking. The prohibition of intentional spinning in the aircraft of concern combined with normally sound spin prevention awareness set you up to be surprised by your spin to which you’ll likely have inadequate thought for recovery as any recovery experience has been based on benignly recovering airplanes and gliders.
Now consider we’re in the spin in the Pitts. Note in the video referenced by Juan the spin is left with that single engine tractor propeller. Recall your left turning tendencies? Yep, power adds left yaw. So of course the spin gets more yaw dominant with the single engine left spin. More yaw flattens. But what of the right spin? Then those tendencies help reduce yaw. Less yaw tends to steepen. So Juan’s absolutes are not absolute. Yet some but not all upright right spin light aircraft may initially steepen with power yet due to inadequate aerodynamic forces end up flatter as the faster rotation sets the masses’ centrifugal forces toward flattening. USAF TPS gives a pitch discussion which they later translate to power in a right spin:
As you would expect, applying nose down elevator increases the downward aero moment. If this aero moment is sufficient to reduce the angle of attack below stall, you recover from the spin. If it isn’t, you may stabilize in a different spin mode. And not only might you stabilize at a different spin mode, it will probably be a faster spin mode. One way the aircraft can balance the higher nose down aero moment is by rotating faster to create a higher nose up inertial moment. The material we’re covering in this course is insufficient to predict how an aircraft will migrate from one spin mode to another, but the resulting spin after applying forward stick might be at both a higher spin rate and also at a higher angle of attack!(15)
Just like the case of insufficient elevator power, the small increase in nose down pitch acceleration may stabilize the aircraft in a faster and possibly flatter new spin mode.(16)
In single engine right spin adding power will pitch down against stall and will reduce yaw but with the pitch down making you rotate faster thus you could lose the improvement against increased centrifugal effects if aero insufficient to counter. In a pinch, it could help you, but it could also be inconsequential or worse. Normal < Fast < Flat < Fast Flat
Confusion of should you or shouldn’t you add power to help as it might hinder… you can see why we like PARE yet start to see the incompleteness of PARE. If PARE hasn’t worked and you’re in a right spin, power could help you. If you have nothing else to try, why wouldn’t you? It might not cancel the yaw but it may enable the rudder to complete the task while it may not solve the stall but may make it easier for the elevator. But it carries risk of accelerating and this has further risk of inertial difficulty. If stuck in a right spin, however, I’m likely to give power a try. When it comes to the twin, trying something if stuck will carry over.
Now consider methods of spin entry. In the Pitts, you likely pulled power and held if not elevated the nose looking to stall while nearing or into the stall you added rudder. You could enter a spin in this manner in a light twin though the common way is to stall under asymmetric power conditions. These entries make for a distinction while conservation of momentum comes into consideration.
Juan was correct to note the mass along the wings of the twin. This mass doesn’t want to rotate. Neither roll nor yaw. It has inertial resistance to both. This airplane initially does not want to spin. Should you pull power to idle and hold the nose up to stall and add rudder, those wings are going to are going to be sluggish relative to that Pitts spin. And they’re going to be sluggish relative to the asymmetric power entry. They will yaw with all mass, wings and fuselage, which may lead you to think it would want to be flatter due to centrifugal forces. Yet it will still have rolling moments as the rudder is above the center of gravity and asymmetric lift will still play with retreating wing versus advancing wing. The entry will be flatter and more sluggish in a relative sense compared to the snap of asymmetric power. Yet it will still go over the top and go steep. If it catches a steep spin, the wing mass will satisfy centrifugal pushes to make it perpendicular to the axis of rotation. The wing mass would be perpendicular in a hypothetical (non-existent) nose completely down rolling spin just as it is should it attain a flat spin. Only fuselage mass changes magnitude of contribution between the two extremes of spin. Wing loaded go steep as seen in videos.(17) Once the spin starts, the wing loaded plane wants to go spinning.
The prevalent spin entry for the twin involves losing lift while the power is unbalanced. This creates a significant aerodynamic influence. Those propellers in front of the wings throw lots of air over the wings increasing their lift. If you have one good engine while the other is feathered, or worse, windmilling with no power hence reducing airflow, you’ll have that much better lift over the good engine outside advancing wing side. When you stall, the inside wing not only will stall more, it will stall first hence you will get a significant roll which that inertia will not stop. This is why the light twin will “go over the top” quickly. This is also why you want to be aggressive at pulling both power levers to idle and pushing the yoke to break AOA.
Yet Juan was wrong when he said “If you allow that aircraft to stall before you conclude your Vmca demo, you’re gonna find yourself right over the top in a flat spin right away.” You’ll either be flat or you’ll go over the top, but not both; these are mutually exclusive events. Over the top not flat spin entry is associated with asymmetric power into stall. Another influence after spin entry must change the spin mode to flat for the asymmetric power situation to go flat. I’ll concede sustained power on the outside can be such an influence. Fuselage loaded aircraft like to spin flat, not wing loaded. Yet wing loaded can transition to a flat mode. If they do then their inertias tend to lock them into it. They don’t like to enter the flat spin but they like to stay in it if they get there.(18) Juan’s error here is in ‘flat’ paired to ‘right away.’
Imagine two barbells crossing their handles at your cg. Alter their weights and/or lengths and you can change wing load versus fuselage load. Note how little inertia to roll the fuselage load has compared to wing loaded meaning it will be both easier to roll and easier to stop the roll. Yet its yaw will be hard to start and hard to stop while yaw flattens spins. The wing loaded can roll steep with no issue, should that roll start with little yaw, the empennage will still weathervane while the nose will already have dropped in the initial spin roll. It is only if something else drives the four masses flat in which it will go flat but once flat all four will work to keep it flat. Centrifugal forces drive spinning objects perpendicular to their axis of rotation while flat spins have their axes of rotation along their vertical axes. Were you to be able to have a pure roll spin, it would align with the longitudinal axis.
How might one wrongly drive the light twin flat? By not immediately pulling power. Look at P38 Vmc Roll No Drill(19) times 8:10–8:36 vs 3:40–5:15. Initial spin will be steep though we’ll see in a moment how outside engine power can subsequently drive this flat.
Regarding wing loaded planes like the light twin, Mason has this to say,
A wing-heavy airplane, or an airplane that has considerable wing mass in the form of structure, fuel, engines, etc., may develop considerable rolling energy about the roll axis once a spin develops. If the nose is pitched upward about the pitch axis, the precessional effect of the rolling mass will result in a pro-spin yaw.(20)
Previously we mentioned flywheels. The masses create a flywheel in yaw and a flywheel in roll. Yet these flywheels can differ in mass. And there is a third for pitch. All will influence yet yaw will dominate. The previously mentioned aerodynamics are still at play, all forces accumulate some constructive some destructive. Start thinking about your gyroscopic concerns with rigidity in space, which is the spinning thing’s way of saying conservation of momentum, and precession.
With a spinning object, a force applied propagates ninety degrees ahead of the spin. This is another way of visualizing that the spinning object repositions towards aligning the rotation with the force. Note also if the force should be in alignment with the rotation then it is merely an accelerant or a braking.(21)
Precession is the resultant action, or deflection, of a spinning rotor when a deflecting force is applied to its rim. As can be seen in Figure 5–49, when a force is applied, the resulting force takes effect 90° ahead of and in the direction of rotation.(22)
Precession is the tilting or turning of a gyro in response to a deflective force. The reaction to this force does not occur at the point at which it was applied; rather, it occurs at a point that is 90° later in the direction of rotation. This principle allows the gyro to determine a rate of turn by sensing the amount of pressure created by a change in direction. The rate at which the gyro precesses is inversely proportional to the speed of the rotor and proportional to the deflective force.Using the example of the bicycle, precession acts on the wheels in order to allow the bicycle to turn. While riding at normal speed, it is not necessary to turn the handle bars in the direction of the desired turn. A rider simply leans in the direction that he or she wishes to go. Since the wheels are rotating in a clockwise direction when viewed from the right side of the bicycle, if a rider leans to the left, a force is applied to the top of the wheel to the left. The force actually acts 90° in the direction of rotation, which has the effect of applying a force to the front of the tire, causing the bicycle to move to the left. There is a need to turn the handlebars at low speeds because of the instability of the slowly turning gyros and also to increase the rate of turn. Precession can also create some minor errors in some instruments. [Figure 8–19] Precession can cause a freely spinning gyro to become displaced from its intended plane of rotation through bearing friction, etc. Certain instruments may require corrective realignment during flight, such as the heading indicator.(23)
Now that we’re looking mass distribution and inertia, we need to look again at the effects of ailerons.
Darrol Stinton(24) derives imposed inertia moments for us as:
Roll = L = (Bq)r — (Cr)q = (B-C)qr [& as C~=A+B, L ~= -Aqr]
Pitch = M = (Cr)p — (Ap)r = (C-A)pr [& as C~=A+B, M ~= Bpr]
Yaw = N = (Ap)q — (Bq)p = (A-B)pq
in which
A = Ixx = wing inertia
B = Iyy = fuselage inertia
C = A + B = Izz = yaw inertia
p = roll rate
q = pitch rate
r = yaw rate
With the wing loaded airplane or glider, C>A>B
& the fuselage loaded airplane C>B>A
while C is dominant in all; this is to say the yaw flywheel is the largest though roll’s and pitch’s trade at being the middle or smallest depending on loading.
Of biggest interest to us is the yaw flywheel to which we see pitch and roll influence. How these influence depends on our mass distribution.
A = B in the equally loaded or light loaded aircraft, driving (A-B) to zero thus leaving aerodynamic forces dominant. In the light loaded aircraft, neither roll nor pitch moments help much with yaw moments. Though, as we saw, roll surfaces can create asymmetric drag therefore yaw aerodynamically.
A > B is the wing loaded aircraft. Aileron away reduces on p reducing yaw N.
Slow power reduction of the light twin such as to have sustained yaw from the outside engine can drive the plane into a flat spin as imposed roll and yaw moments contribute to pitch, they’re essentially multiplied by the fuselage portion of mass, which is small. The light twin will have significant immediate yaw and, driven by increased engine airflow over outboard wing, large roll which will create a flattening influence but such will not be immediate. Think pitch up = small mass times big yaw times big roll. The plane will “go over the top” and steep before going flat. As most mass is in the wings, centrifugal forces could lock the steep with the plane of rotation perpendicular to axis of spin. This is to say something needs to drive you flat in order for flat to be locked centrifugally. Sustained pro-spin yaw from the outside engine not at idle would be such a thing increasing positive pitch moment. There is an amplifying feedback at play as the pitch up increases pro-spin yaw while reducing roll thus further flattening. Perhaps this is why Juan says “If you allow that aircraft to stall before you conclude your Vmca demo, you’re gonna find yourself right over the top in a flat spin right away.” Yet ‘right away’ is misleading. This does, however, give value to Juan’s and Scott Perdue’s point in BSWorks-5 Flat Spins and Twins(25) regarding how we should treat spin prevention during departure as a separate and immediate emergency from spin recovery.(26) Time 10:30,
It’s not a spin recovery
Time 7:26,
The prevent is the part that you don’t actually develop the spin or go more than one turn, it’s less than half one turn, you unload the airplane, use coordinated rudder and aileron to stop the use and roll out of the bank and then recover
A < B is the fuselage loaded aircraft. The sign on N becomes negative therefore we want to increase p via aileron into the spin.
It is with these that we will attempt
To win the game of spin-recovery one must use the flying controls to alter the equilibrium condition and induce rotary inertias to collaborate in producing ‘out-spin yaw’.(27)
Yaw = (A-B)pq
With the wing loaded smaller B airplane or glider, in spin roll gives increases in spin yaw as A-B is positive. Roll away, however, and yaw is decreased giving rudder a better chance to defeat it. Note: if inverted, you get a sign change on p.
Stinton(28) presuming an upright left spin:
Ailerons act in the normal sense*, i.e. stick to the left produces a rolling moment to the left. This changes the wing tilt. When B>A and the fuselage inertia is dominant, applying in-spin stick (starboard aileron DOWN in a spin to the left), raises the starboard wing above the horizon. Yaw and roll are coupled and some of the rotary momentum is converted into roll, the enhanced rate of rotation, acting on the fuselage masses, produces out-spin yaw which favours recovery. Not only that, but down-aileron on the wing farthest from the spinning axis causes drag which slows spin rate and generates a component of out-spin yaw.
When A>B and the wing is dominant, in-spin stick couples roll with yaw. The rate increases and the masses in the wings cause yaw in-spin as they rotate about the Z-axis, in an attempt to take up a common plane of rotation. Simultaneously the spin tends to flatten. Out-spin stick has the opposite effect, decoupling roll from yaw. The radii of gyration of each wing-mass increase, reducing the spin rate and angle of attack, with the nose going down, assisting recovery.
…
Forward movement of the stick, elevator-DOWN, opposes the tendency of the fore and aft dumb-bell masses to seek a common plane of rotation, by displacing them vertically in Fig. 12.4a(2). This, by reducing their radii of gyration about the spin axis, causes an increase in rotation like a dancer or a skater speeding up in a pirouette by drawing in the arms. What happens next depends on the ratio of B/A. When B>A [fuselage loaded] the effect is pro-spin — the nose tends to rise and the spin to flatten. When A>B [wing loaded] elevator-DOWN is anti-spin and assists recovery
…
Rudder is the primary source of out-spin yaw. The moment which is produced alters the wing-tilt, raising the outer wing in the spin above the horizon. The effect is anti-spin when B>A [fuselage loaded] and pro-spin when A>B [wing loaded].
Note despite rudder inertial, its aero dominates hence always opposite yaw opposite turn needle for out-spin. The wing loaded pro-spin inertia simply means you’re just getting a little less total out-spin yaw from rudder, but you’re still getting out-spin with it away from the spin.
As we can see, inertially speaking, ailerons help in spin recovery though direction depends on airframe loading.
* recall that the USAF TPS noted the ailerons in the spin had the downward aileron trapping a pocket of air while the upward aileron let some air slip hence how we get to normal sensing here.
The flywheels can be seen on Stinton page 518:
Stinton pages 524 & 525:
All of this can be boiled down to a heuristic or rule of thumb (ROT) provided by the USAF TPS.(29)
With this I call foul to Juan’s admonition “never aileron.”
Though a fuselage loaded case, the T-45 Goshawk demands lateral stick input in its spin recovery(30):
In this same OCF T-45C FTI, the USN calls out lateral stick for most fuselage aircraft though neutral for wing loaded and light loaded yet these wing loaded don’t seem as relatively wing loaded to fuselage lack of load as the light twin.(31) These platforms likely have better elevator capacity than the light twin too. Mason wrote similarly of the T-38:
For a fuselage-heavy loading, the aileron is the primary recovery control. Fighter aircraft with short wings and long, heavy fuselages respond best to this control. … ailerons used in the direction of the spin are essential to stopping the spin of airplanes with heavily loaded fuselages, such as the T-38.(32)
Juan told us “Why don’t you add power opposite to the spin in the multi-engine aircraft? NO! Anytime you add power to an engine, you’re going to flatten the spin.” This is only conditionally true which is to say that the statement is false. You cannot correctly make conditional statements in absolutist terms. In so doing you falsify the entire statement. Context matters.
Consider per the USAF TPS such is part of the F-15E recovery.(33) And it is of tertiary concern in the generic spin recovery sense.(34)
We should note that thrust difference to counter yaw is used last as such could put you into a progressive spin meaning you correct the spin the one way to merely have stall and yaw the other way hence now spin the other way. Be slow to put it in and quick to take it out!
As the F-15E uses asymmetric power against the yaw of a spin, it obviously cannot be “never.” Similarly in the light twin, we could use power opposite the yaw. In this case we’d have the benefit of roll against it too due to increased airflow over the inside more stalled wing. Such should be seen as last ditch attempt, however, as this excess yaw should it break the spin will subsequently be yaw at or near stalled condition setting up a spin the other way (progressive spin). Yet to discount the physics here may cost you your life. Per Mason writing to single engines,
As a rule, power increases rotational speeds and makes spin recoveries more difficult. Normally, the first step in a flat spin recovery is to close the throttle. However, if everything else fails, it may be worth a try. During a right spin, the precessional forces due to the propeller would tend to lower the nose (if they are strong enough to overcome other effects), and this could possibly be a method of breaking the stability of the spin and starting a recovery. I wouldn’t count on the use of power being helpful however, it has been my experience that it only speeds up the rotation.(35)
And Mason writing to twins,
If power is available (engines tend to quit during flat spins), it may be worth trying asymmetric power against the direction of rotation (left spin, left throttle)(36)
Note, I’m not copying Mason’s full light twin recommendations here as I believe he leaves a gap between not recovering and flat spin. He discusses flat though not what to do if not flat and not working. Instead, I’ll provide my own general versions for wing loaded and fuselage loaded below incorporating the USAF TPS ROT. Then I’ll adapt the wing loaded from an initial PARE through the ROT to match the Beech Baron’s and Juan’s generalized twin recommendation appending the ROT with flat as its own concern from Mason after this ROT.
Consider spin testing often uses a sort of safety device in the even the spin cannot be recovered. This is typically a drag parachute. Though rockets are an alternate means. What are rockets if not asymmetric thrust?
Stinton,
The anti-spin rocket, fitted to both wing tips and/or tail can be effective. It is less popular than the mechanically simpler anti-spin parachute, because it combines explosives with electrics, and still needs safety devices.(37)
Mason,
NASA has been experimenting with hydrogen peroxide wing-tip rockets that control roll and yaw. They are used to change the roll and yaw patterns of the spin and have been used more than the spin chute to stop unrecoverable spins.(38)
USAF TPS(39),
Though they discourage the use due to chance of mistake, the 1936 NACA Technical Note 555 says ailerons could help as could power,
The advisability of using ailerons as an additional means of recovery is debatable. It is suggested, however, that the use of ailerons relative to the individual airplane be studied, inasmuch as on some airplanes they aid appreciably in recovery, provided that the proper displacement is used. The proper displacement may be either against or with the spin, depending on the particular airplane.(40)
Use of the throttle in an attempt to recover from a bad spin, although effective at times, is very poor practice and generally should be considered as a measure to be tried only as a last resort.(41)
While the NACA note says ailerons are debatable and that power is poor practice, this means don’t use initially; it does not mean, however, never use. It says they may be effective meaning such is highly circumstantial. If you’re at the end of your procedures and you’re at the end of your wits yet you’re stuck in the spin, perhaps you ought to consider trying them.
I need to thank Catherine Cavagnaro(42) for bringing the NACA note to my attention in her spin presentation and I like that throughout her presentation she emphasizes that with spins, it depends. Each platform is different. She shies away from aileron but she does not discount such. She discusses what happens in her plane showing it can be rather helpful or slightly hurtful, time 30:40:
I spun it [C-152] to the left and gave it full aileron to the left so full aileron into the spin, and what happened is fascinating. When you give in this airplane in this platform when you put ailerons into the spin, it makes it go a little flatter, flatter being bad, our nose is pitched up a little higher… ailerons into the spin in this airplane are bad, they help hinder the recovery. If you give ailerons opposite the spin in this airplane, you can actually steepen it to the point of recovery.
I’d be curious in this case, is the flatter due to aerodynamic reverse-sense reducing the stall on the left? If so, wouldn’t this also reduce drag hence reduce yaw to the left? Does such give that mostly blanked rudder of yours a chance? Since you’re willing to play with characteristics, might be interesting to see pro-spin aileron, anti-spin rudder while holding pro-stall aft yoke. And then, if no recovery, ease the yoke as opposed to forward seeing if it breaks free with less forward than normal. Though this could alternately be a case of pro-spin normal sense aileron accelerating the roll spin with aerodynamics unable to counter centrifugal forces hence flattening, in which case these test methods would not work. (USAF TPS noted air pocket & air slippage into USAF TPS noted spinning faster can go flatter by virtue of being faster)
Catherine warns of a similar aileron good in hers potentially being catastrophic in a different plane though I think she misses a factor in this concern, time 31:50:
with a PA-28*, though, this is based on work done at the University of Tennessee Air and Space Institute where a dynamically scaled model was spun and when ailerons were put against the rotation against the spin it made it go flat to the point of unrecoverable no matter what they did.
With such, I start to question the scaled model of PA-28. How was the weight? How was the cg? What about bottom surface area as exposed to the relative wind in a flat spin? That reduced relative weight in front get held up by lots of area forward of cg?
* This was a scaled model, not the airplane. Stinton reminds us of the Squared-Cube Law(43):
Over-control of masses in motion can be a problem with small and very light aero planes. Such aircraft (scale replicas, small home built, microlights, and ultralights) are dominated by their aerodynamics. Inertia plays the lesser part in their handling qualities. The ‘square-cube law’ asserts itself, in that aerodynamic forces are dependent upon area, i.e. length squared, while mass depends upon density of the material and the volume which it occupies, in short, upon length cubed.
aerodynamic force / inertia varies as (length^2 / length^3) = 1 / length
(aerodynamics / inertias) vary as 1 / scale)
as aircraft are built smaller, their aerodynamics increasingly dominate their inertias and they become quick-reacting and over-lively.
Squared-Cube tells us aero becomes more dominant in the balance between aero and inertia as we get smaller, but squared-cube plays within aero itself as mass forward of cg is part of the weathervaning of an airplane, not just surface area aft of cg. Surface area forward of cg plays too while this surface area could generate force sufficient to hold up the reduced forward mass of the scaled model while not necessarily having the same full scale. This is to say models are starting points for exploring spin characteristics, not ends. Less mass, more sail.
Further, NACA Research Memo L57F12(44) yields scale model concerns for air flow & friction with Reynolds number differences,
This effect could lead to a variation in the balance of rolling moments and an accompanying difference in wing tilt in the spin. The magnitude of this effect would be dependent on wing section, the magnitude being greater as wing thickness and camber are increased. The difference in wing tilt could, in turn, lead to a difference in the gyroscopic yawing moments (Ix — Iy)pq in the spin. In some instances, the Reynolds number effects may tend to nullify one another — for example, an increased nose-up moment on the model may tend to cause the inner wing to be depressed, whereas a decreased lift on the outer wing may tend to cause the outer wing to be depressed… This could possibly lead to optimistic results in the tunnel for designs having their mass distributed chiefly along the wings but to pessimistic tunnel results when the mass is distributed chiefly along the fuselage.
A PA-28 while light loaded will be more fuselage than wing, hence pessimistic, and it does have thicker more chambered wings, hence greater effect susceptibility.
Though a cautious person would argue that inertia is typically the driver to flat so reducing inertia via scaling down should suggest inadequate aero. While the cautious person says to avoid ailerons in the PA-28 due to wrong ailerons possibly driving the spin flat, one cannot say the airplane has this unrecoverable mode based on a scaled model. Wise to treat it as if it does but cannot say that it does. If you happen to have a spin recovery chute, we could test.
If you have functional recovery procedures, use them. Again, if you’re stuck after executing procedure, maybe it is time to try something different. If what you try starts to flatten, take it out before it has time to complete. If it steepens, good, less yaw more roll and more airflow over both elevators and rudder. Give your procedures some time but don’t become actively resigned.
34:00:
In my airplane, the rudder is not terribly effective…
this is because you’re in an erect spin with low mounted horizontal stabilizers blanking it. Were you in an inverted spin, you’d have plenty of rudder available which may be good as you may not have as much elevator authority or elevator deflection going full aft yoke to break the stall. So, inverted, nice to have rudder to break yaw.
Paraphrasing Catherine,
in my airplane there’s no reason to wait after rudder. In another airplane there may be reason to wait to stop the rotation before pitching forward.
There can and often is an in between giving rudder enough time to reduce yaw thus likely getting to a slightly steeper mode though still spinning such that elevator has an easier time. Consider Stinton with his “standard method”(45) which we would call “British method” and is also what we essentially would do in the T-37.
- Throttle CLOSED.
- Ailerons NEUTRAL.
- CHECK that you are in a spin, not a spiral, and also the DIRECTION of rotation.
- Stick BACK (i.e. conventional elevator- UP).
- Rudder FULL against the indicated direction of turn.
- PAUSE (say, long enough to count one hundred-two hundred-three hundred) allowing the rudder to bite and take effect. THEN:
- Move the stick progressively FORWARD (elevator NOSE-DOWN) until rotation stops.
- EASE OUT of the ensuing dive.
Mason makes comment here too,
The rudder-followed-by-stick sequence is important. Premature use of down elevator tends to deflect airflow away from the vertical fin and rudder, making these surfaces ineffective in damping and stopping rotation. Also, premature use of down elevator may result in translating yaw into roll, increasing the speed of rotation. When this occurs, the rotational radius decreases, transmitting the energy into a smaller circle. An ice skater performing a pirouette utilizes this same principle. By pulling the arms inward, the skater moves the mass to a smaller circle and thereby increases rotational speed.(46)
P-38 Lightning Spin Recovery Training(47) discussing an obviously wing loaded aircraft clearly calls for a pause after throttles idle and rudder opposite yaw before elevator control column forward (but only toward neutral position (despite what looks visually to me to be beyond neutral)). In theirs, they say to wait half a turn before doing so.
Catherine(48) does take a stab at fuselage and wing loaded aircraft though I think this a little off, 51:30:
fuselage loaded respond well with rudder,
not really. The example was really a light loaded plane. We’re seeing more via aileron and split stabilator with heavily fuselage loaded aircraft.
Wing loaded respond well to elevator.
Eh, yes, but this does not discount other surfaces. Yes, USAF TPS agrees to wing loaded really using elevator. But they also share a report for the USAF Academy on the ASK-21 glider(49). Gliders are wing loaded too. The ASK responds very well to rudder. It also responds to aileron.
The aileron effects on the spin characteristics of an aircraft are generally well documented in spin theory literature, and the ASK-21 test results were typical of a wing loaded aircraft design. A detailed discussion of aileron effects on spin characteristics is contained in Appendix H.
For the ASK-21 spin tests, lateral stick against the spin achieved a noticeable bank angle away from the spin as well as a nose down pitch rate. Most of these spins resulted in recovery as the yaw rate decreased, roll rate increased, and the nose pitched down leaving the aircraft in a steep sideslip to terminate the spin. In a few cases, the aircraft remained in a spin with the bank angle away from the spin direction. Therefore, lateral stick against spin was not a reliable contributor to spin recovery.
Lateral stick with the spin increased rotation rate, but this effect was masked by the oscillatory characteristics of the spin. In the majority of tests flown, lateral stick into the spin achieved a slightly higher rotation rate and a more sustainable spin. The results of testing isolated lateral stick inputs indicated that neutral stick was the best position for recovery(50)
Were rudder to be insufficient, one can see with wing loaded aircraft, aileron assist away from the spin, the same direction as the rudder, may help. Note, if inverted, aileron into the spin as the upside-down wings mean such will actually act as aileron away.
During spins, ailerons remained effective in producing a bank angle change in the proper sense. By using ailerons to reorient the aircraft attitude on the spin axis, a component of the spin rate vector, W->, can be generated on the y body axis (lateral axis), creating a pitch rate q.
Pitch rate caused the aircraft inertial moments to affect roll and yaw accelerations. This can be seen from the inertia terms of the yaw acceleration equation of motion.
For the ASK-21, Izz>Ixx>Iyy. Therefore, ailerons against the spin produced antispin yaw acceleration. Conversely, ailerons with the spin produced prospin yaw acceleration.(51)
More to the point, multiple surfaces may need to work cumulatively. Each chips away a bit at yaw and/or stall. For the wing loaded, you may very well want elevator, rudder, and aileron all working for you. Repeating Catherine’s point, each airplane is different. Mason wrote so too
what works on one airplane may not work on another. Some airplanes respond best to rudder during spin recoveries, others require a forceful application of forward stick after the rudder is applied. Some airplanes will spin faster if back stick is released or pushed forward too soon, others will stop spinning as soon as the back pressure being held against the stick is released. A few airplanes stop spinning faster with the application of aileron opposite to spin direction (not recommended in most airplanes). Airplanes that are exceedingly spin resistant may require the use of aileron opposite to rotation to induce and continue a spin. The same procedure may result in flat, unrecoverable spins in some airplanes.(52)
In Catherine’s presentation, she gets into a discussion with an audience member regarding a spin recovery in a Mooney at time 56:50. This is an interesting discussion to which it is worth listening. To summarize it, spins are prohibited in the Mooney as the Mooney likes to transition to a flat spin mode from which recovery is difficult. The gentleman making a comment told of inadvertently getting into such a spin to which Catherine quipped, “what, did you climb on the glare shield?” both noting somehow the person commenting survived and emphasizing a point that forward cg is better, shifting more forward can help. Apparently, the gentleman in question did shift the cg though he did so dynamically having both himself and the other person in the plane slide their seats aft and forward in phase with each other trying to rock the cg swinging the balance of the plane ultimately going forward. Such did pitch the plane down out of the flat spin for a near the ground near death recovery. I really like this story as they had exhausted their spin recovery procedure yet did not merely sit at the end. Instead they engaged rather than resigned. With this I’d also suggest looking into Chaos and Complexity in Dave Snowden’s Cynefin. (Cynefin in aviation also makes me think Safety Differently with Sidney Dekker, Human Performance with Todd Conklin, Human & Organizational Performance with Bob Edwards, more Cynefin with Gary Wong, John Boyd in general, and thoughts for Danny Kahneman and Gary Klein regarding decision making. Scott Page has his complexity lectures too.)
It does not matter if you “have entered test pilot territory,” if you’re there, you are there and have no choice about it. Accept it and move on. Do something. Try something. If you have means to get out, do so. If not, keep trying things.
What would I recommend for any given airplane or glider? Start with PARE then add as needed. For the Light Twin, I’ll start with the Baron’s POH procedure as Juan showed cross checking Mason’s while adding the extra concerns to the end yet holding Mason’s concerns if flat to a separate flat generic procedure. Before these, however, we should take Scott Perdue’s point to consider Spin Prevention* first. Note spin prevention is not the same as spin avoidance. Spin avoidance is before any portion of the spin. Spin prevention is for the spin entry and into incipient spin preventing developed spinning. Spin recovery is for already in incipient spin and in developed stabilized spin.
Spin Prevention.
Immediately upon us commanded roll and/or yaw
- Stick (or Yoke) — UNLOAD
- Power(s) — IDLE
- Rudder — AGAINST ROLL OR YAW
* I need to thank Scott Perdue and Juan Browne for explicitly bringing out the prevent. It is something I would intrinsically do, left wing drop hit right rudder and push nose, but it is not something I would verbalize and teach. They’re making the implicit explicit and in this case that is a good thing. Previously I would merely teach avoidance and recovery, now I will also teach this in between prevent.
For the Wing Loaded airplane and glider spin:
- Power(s) — IDLE
- Ailerons — NEUTRAL
- Rudder — OPPOSITE YAW (Opposite Turn Needle) and PAUSE
- Elevator — STICK (or Yoke) FORWARD (Aft if Inverted)
- Ailerons — Upright OPPOSITE YAW (With Rudder); Inverted WITH YAW (Opposite Rudder)
- PAUSE
If still spinning after three turns,
- cg — SHIFT FORWARD (as able slide seats forward, throw stuff forward, dump any aft ballast)
- Elevator — TRY ROCKING AFT AND FORWARD A FEW TIMES (Ending Forward (End Aft if Inverted))
- Power -
- TRY IMPULSE only on Engine Inside Spin (Turn Needle Side) if Multi-Engine
- TRY IMPULSE if high mounted engine(s)
- TRY IMPULSE if in Right Yaw Spin & Single Engine
- Power -
- TRY SLOWLY RAMPING only on Engine Inside Spin (Turn Needle Side) if Multi-Engine
- TRY SLOWLY RAMPING if high mounted engine(s)
- TRY SLOWLY RAMPING if in Right Yaw Spin & Single Engine
Rotation Stops,
- Power(s) — IDLE
- Rudder & Ailerons — SMOOTHLY NEUTRAL
- Recover from Dive
- By ‘impulse’ I mean quickly advance and immediately retreat back to idle.
- By ‘ramp’ I mean slowly advance power stopping the advance as soon as effect is perceived.
- A single engine airplane is unlikely to be wing loaded though it is possible. Perhaps you would treat a motor glider as such. Full wingtip fuel tanks might drive you to be such.
- Power [Levers] & Throttles used interchangeably
- Upright means positive g High AOA; Inverted means negative g Low AOA; you may have your belly to the sky yet you’re in an upright spin
For the Light Twin airplane:
Perform Step One pieces as near simultaneous as possible with Throttles Closed taking primacy:
- Powers — IDLE, Rudder — OPPOSITE YAW (Opposite Turn Needle), Yoke — FULL FORWARD
- Ailerons — NEUTRAL
- PAUSE
- Ailerons — Upright OPPOSITE YAW (With Rudder); Inverted WITH YAW (Opposite Rudder)
- PAUSE
- cg — SHIFT FORWARD (as able slide seats forward, throw stuff forward)
- Power — TRY IMPULSE only on Engine Inside Spin (Turn Needle Side; Opposite Rudder)
- Power — TRY SLOWLY RAMPING only on Engine Inside Spin (Turn Needle Side; Opposite Rudder)
Rotation Stops,
- Power(s) — IDLE
- Rudder & Ailerons — SMOOTHLY NEUTRAL
- Recover from Dive
- By ‘impulse’ I mean quickly advance and immediately retreat back to idle.
- By ‘ramp’ I mean slowly advance power stopping the advance as soon as effect is perceived.
- Power [Levers] & Throttles used interchangeably
- Upright means positive g High AOA; Inverted means negative g Low AOA; you may have your belly to the sky yet you’re in an upright spin
NASA Technical Note D-6575 confirms inside engine can be useful,
For asymmetric power for a twin-engine configuration with the engines mounted on the wings, power from only one engine can produce a large asymmetric yawing moment, which will be favorable or adverse to the spin and recovery, depending on the direction of the moment. Both model and full-scale spin-test results of multiengine airplane designs have shown that power on the outboard engine (e.g., the right engine in a left spin) can create a large prospin yawing moment, which can cause a flatter and faster spin. On the other hand, power on the inboard engine can create an antispin moment to aid spin recovery. Normally, the manipulation of thrust can be confusing and disastrous if the power is applied to the wrong engine. Therefore, unless asymmetric power is necessary to aid recovery, it is generally recommended that for a multiengine airplane, the throttle be retarded to the idle position on all engines during a spin.(53)
We saw earlier that outside power leads directly to yaw and roll while creating pitch up and more roll while both roll and yaw drive pitch up, which in turn converts roll to yaw in an amplifying feedback thus driving flatter faster. Other factors that could catch this feedback are delayed stick forward and pro-spin aileron (roll into the spin). Leaving the stick aft (erect spin) gives pitch up which converts roll to yaw and flattens. With these, step one as well as the prevent should be emphasized.
For the Fuselage Loaded airplane spin (light loaded including most single engine GA can also use):
- Power(s) — IDLE
- Ailerons — NEUTRAL
- Rudder — OPPOSITE YAW (Opposite Turn Needle) and PAUSE
- Elevator — STICK (or Yoke) NEUTRAL
- Ailerons — Upright WITH YAW (Opposite Rudder); Inverted OPPOSITE YAW (WITH Rudder)
- PAUSE
If still spinning after three turns,
- Elevator — Erect AFT Inverted Forward (anti-spin inertial (reduces yaw))
- Pause
If no change
- Elevator — Erect FORWARD Inverted AFT (anti-stall aero)
- cg — SHIFT FORWARD (as able slide seats forward, throw stuff forward, dump any aft ballast)
- Elevator — TRY ROCKING AFT AND FORWARD A FEW TIMES (Ending Forward (End Aft if Inverted))
- Power -
- TRY IMPULSE only on Engine Inside Spin (Turn Needle Side) if Multi-Engine
- TRY IMPULSE if high mounted engine(s)
- TRY IMPULSE if in Right Yaw Spin & Single Engine
- Power -
- TRY SLOWLY RAMPING only on Engine Inside Spin (Turn Needle Side) if Multi-Engine
- TRY SLOWLY RAMPING if high mounted engine(s)
- TRY SLOWLY RAMPING if in Right Yaw Spin & Single Engine
Rotation Stops,
- Power(s) — IDLE
- Rudder & Ailerons — SMOOTHLY NEUTRAL
- Recover from Dive
- By ‘impulse’ I mean quickly advance and immediately retreat back to idle.
- By ‘ramp’ I mean slowly advance power stopping the advance as soon as effect is perceived.
- Power [Levers] & Throttles used interchangeably
- Upright means positive g High AOA; Inverted means negative g Low AOA; you may have your belly to the sky yet you’re in an upright spin
Applying to your aircraft, determine whether you are wing or fuselage loaded, then add a pause then add aileron then then cg shifting then power steps after the end of your prescribed procedures. For a spin rated plane, the Pilot Operating Handbook (POH) is master, but for those not spin tested like light twins(54), maybe consider the POH with some salt. How can a test pilot give the best procedure for a specific plane should the plane never have been tested?
Note engine position may not merely be lateral, engines may also impact pitch. Low slung engines gonna pitch up, high slung engines gonna pitch down. Single engine tractor propellers may increase the down force of the empennage meaning pitch up (think trim stall), more likely with lower mounted horizontal stabs (yet gyroscopics, left spin power flattens, right spin pitches down but may have secondary effects). T tails may avoid much of such aero (but not gyroscopics), but no promises.
Going Flat: The flat spin is a special situation that is difficult to exit as all masses are coupled in the rotationally sought plane perpendicular to the axis of rotation. The yaw flywheel has become the only flywheel meanwhile aerodynamics leave us little influence as the empennage only has air from below not from in front while below air hits forward fuselage too balancing against this aft below air. Any weathervaning tendency is reduced. Mason gives plenty of discussion regarding flat spin recovery,
Because the use of aileron into the spin direction increases rotational speed, it is possible for the centrifugal force to become so strong that it would tend to flatten the spin.* However, one of the recommended ways of recovering from flat spins in some airplanes is by using fully deflected aileron in the direction of the spin. This procedure utilizes the drag of the depressed aileron to resist autorotation and to lift the outside wing. The lifting tends to recouple roll with yaw to restore a normal spin pattern. Also, the raised aileron on the inside wing will serve as a spoiler to the lift and autorotational force being contributed by this retreating wing.(55)
* as we’re already in a flat spin, we’re not worried about accelerating the spin such that inertias flatten us.
If the airplane does not respond to the controls and appears to be spinning flat (within 45dg of the horizon), apply full aileron in the direction of the spin. This action will tend to spoil the lift on the retreating wing and reduce its autorotational contribution.(56)
Thus we can take for the flat spin
- PARE — IDLE, NEUTRAL, OPPOSITE, FORWARD (Erect) AFT (Inverted)
- Ailerons — Upright INTO YAW (With Turn Needle, Opposite Rudder), Inverted AWAY FROM YAW (Opposite Turn Needle, With Rudder)
- SHIFT cg FORWARD
- Elevator — SWAP (better inertial if fuselage loaded)
- Pause
- Elevator — SWAP (aero down, better inertial if wing loaded)
- Ailerons — CYCLE
- Power — IMPULSE INSIDE ENGINE (Increase Turn NEEDLE SIDE)
- Power — RAMP INSIDE ENGINE
- Power Single Engine & Right Spin — TRY IMPULSE THEN RAMP
Be it flat spin or simply stuck spin, if you can shift the cg forward, it will help. Can you shift seats forward? Throw stuff forward? Try rocking the seats as did the Mooney pilot? Try rocking the pitch always ending forward (aft if inverted)? Try also rocking the ailerons always ending into the spin (away if inverted). If not working, try asymmetric power against yaw (power up the inside engine). Keep fighting.
In spin recoveries, we have yet another concern raised by Mason applicable to all airplanes and gliders in terms of how do we determine in which direction are we spinning. The direction of yaw is our direction of spin while roll and yaw match when erect but go opposite if inverted. A back of airplane view Turn Coordinator may lead you astray while the needle style Turn and Slip Indicator will always serve you well:
Pilots have spun all the way to the ground thinking they were in an erect spin when they were actually inverted. An inverted spin may not appear all that inverted in attitude. The pilot may be only slightly against the belt, perhaps only a little more than an erect, steep, nose-down spin.
During an inverted spin a pilot must be aware of which way the nose is yawing. This is determined by looking directly down the nose. It is also true that the old-fashion turn indicator determines yaw equally well, whether the airplane is erect or inverted. However, the modern turn coordinator requires interpretation. It displays both yaw and roll as a roll. This information could be interpreted as yaw during an erect , but would be extremely confusing during an inverted spin. The ball portion of both these instruments means nothing during a spin.(57)
The USN via the T-6 OCF FTI gives us emphasis for Turn Indicator,
Disorientation experienced by the pilot during an inverted spin is primarily because the yaw and roll occur in opposite directions. Pilots are more sensitive to motion about the longitudinal axis than the vertical axis, and are consequently more likely to interpret an inverted spin in the direction of roll rather than the direction of yaw. Regardless of whether the aircraft is spinning erect or inverted, the turn needle will always deflect fully in the direction of spin and is the only reliable indication of spin direction.(58)
Free Pilot Training(59) gives a review of Turn Coordinator versus Turn and Slip Indicator.
Mason gives us a thought should you be nearing the ground and the spin be unrecoverable while you lack means to eject or bail out. While survivability of a flat spin is low, survivability of a normal or steep spin is even lower. If you’re going to hit, consider trying to drive it flat.
When it comes to the spin, start with your POH. Realize POHs do not actually provide nor was testing done to determine the optimum way to recover the plane. Instead, they give you the optimum way of the tested methods while showing the plane met the recovery criteria of the category of the plane. And this is only done within the modes of the spin to which the testing managed to find the plane entering; it is possible for planes to have additional spin modes. Keep in mind how you may use aileron and if you may use power to assist, when and how. Have a plan for after you complete the procedures should the procedures be inadequate to resolve your situation. If procedure complete and you notice no shift in modes nor oscillations such that you’re just plummeting with no improvement, you may need to try something else. Avoid being actively resigned.
(1) https://www.youtube.com/@blancolirio
(2) https://www.youtube.com/watch?v=n1ZkqWvgIsU
(3) https://www.youtube.com/watch?v=eNdbZ_dvty4
(4) pg. 122, Stalls, Spins, and Safety — Sammy Mason ISBN 0–02–581620–9 https://www.amazon.com/Stalls-Spins-Safety-McGraw-Hill-aviation/dp/0070406960
(5) I say “perceived” as increased media attention and social media may increase awareness without impacting base rate. Though I think Juan’s point that more MEL training is going on may also be impactful. I don’t know which case actually exists.
(6) https://en.wikipedia.org/wiki/Perverse_incentive
(8) see Cynefin and Dave Snowden
(9) https://www.youtube.com/watch?v=v-Bf_Ia4oJA
(10) I have one flight in a Pitts and from it gained an interesting experience. I was flying left wing off another Pitts and I was just a touch “sucked,” behind desired bearing line, and a little rangy. The fix normally would be to add a little power and nudge up and in. So I added power to do this, but I didn’t add enough rudder to compensate. So I ended up even more rangy till I realized this error. Such is a minor power change for which is a Cirrus or Vans, you would not need such rudder change. I say this to show just how susceptible that Pitts is to the four turning forces.
(11) USAF TPS speaker notes slide 20, 2007 Spin Recoveries presentation
(12) pg 1–8, 2012 USN T-6A/B OCF FTI
(13) pg. 59, Stalls, Spins, and Safety — Sammy Mason
(14) pg. 83, Stalls, Spins, and Safety — Sammy Mason
(15) USAF TPS speaker notes slide 10, 2007 Spin Recoveries presentation
(16) USAF TPS speaker notes slide 38, 2007 Spin Recoveries presentation; image of slide 38
(17) DC-3 sky dive spin see time 6:30 https://www.youtube.com/watch?v=pNwGwoqvENY, https://www.youtube.com/watch?v=EFyyLbD-Y7o; power split at entry but idled at entry, see steep https://www.youtube.com/watch?v=8D6-fJdGTac; something has to drive them flat like sustained outside power, pro-spin aileron, up pitch elevator.
(18) It is always A or B; it is never A and B. You can have A then B. With more difficulty, you can have B then A. But it is always A or B never A and B
(19) https://www.youtube.com/watch?v=xk91avtFK1E note I disagree with the conclusions of this video believing they are presumptuous while lacking adequate information to so conclude. My comments to it: “Can you say the fuel was lack of systems knowledge when it could easily have been distraction? Similar for the speed?” & “How can you draw the conclusion of a cavalier attitude? I don’t think there is enough there to do so. I’d also suggest looking into Todd Conklin, Sidney Dekker, and Bob Edwards. Inadequate bureaucratic compliance does not equate cavalier. Nor does compliance equate safety.”
(20) pg 64, Stalls, Spins, and Safety — Sammy Mason
(21) image of slide 21 USAF TPS 2007 Inverted Spins and Gyroscopic Effects presentation
(22) pgs 5–31 & 5–32, FAA Pilot Handbook of Aeronautical Knowledge (PHAK) https://www.faa.gov/regulations_policies/handbooks_manuals/aviation/faa-h-8083-25c.pdf
(23) pgs 8–15 & 8–16, PHA
(24) pg 520, in Flying Qualities and Flight Testing of the Airplane — Darrol Stinton https://www.amazon.com/Flying-Qualities-Testing-Airplane-Education/dp/1563472740
(25) https://www.youtube.com/watch?v=uFzFCAqosUY I’ve got some disagreements with this one as well and commented to the effect, “I have a lot of issue comparing an Extra or a Pitts to the light twin for the spin without adequate contrast. Also, single engine (SE) left spin power up is a flattening input but that does not mean such is true for twins nor is it true SE right spin. ME outboard power creates excess pro spin yaw and due to increased airflow over outboard wing increased roll both of which yield pitch up inertias, flattening in and of itself, while also amplifying yaw while reducing roll via pitch feedback further flattening. But power on the inside engine fights all this so long as it be removed before progressive spin (spinning the other way). F-104 fuselage loaded unlike light twins. As is, I would presume, the F-15E. So their ailerons differ. Yet seems we do have spin testing in the BE-76 Duchess which can be found here on YouTube. If you have to do the Vmc demo, you should do it in the Duchess. It responded well to the prevent and demonstrated ability to recover. Now Sammy Mason from Lockheed will suggest as a last ditch try the asymmetric thrust powering up the inside motor if able. As for ailerons, the difference is wing loaded vs fuselage loaded. Erect wing loaded roll away from yaw and turn needle (with rudder). Erect fuselage loaded, roll into yaw and turn needle (away from rudder). Flip both if inverted as the ailerons are upside-down. Aileron is the primary recovery surface for T-38 and, I believe, T-45 (both fuselage loaded). Now consider also high mounted vs low mounted engine impacts with pitch moments. Now I’m a bit confused with the spin axis on the airplane as depicted in this video. Thought spins were helixes meaning the spin axis is in front of the plane? Or so Darrol Stinton shows me. As you flatten this may pull toward the vertical axis of the plane but then this would be through the cg. As for multiple rolls, the F/A-18 has the same restriction, not more than one roll at a time, and there is a good departure video out there from Pax as to why. Meanwhile, I thought the light F-15 could lock into sustained rolls and continue on just fine. Anyway, everyone should have AOA indicators. Mason, Stinton, and Langewiesche all agree with the holding of the stall and keeping wings upright via rudder as training technique. Emphasis on “you have the rest of your life to be a test pilot” should you find yourself in the situation — this means do something, if the prevent didn’t work, try something else. Light twin upright after prevent failed, PARE into aileron away to throw junk forward to move cg up into inside engine power attempts.” — & — “Also, the standard recovery does not guarantee to convert an inverted spin to an erect spin with “idle neutral aft…” Consider how the ‘reversed recovery,’ that is, pushing your stick forward before rudder, is an accelerating influence in your spin. Yes, some planes recover with this, some planes swap from erect to inverted, and some planes spin faster. Some even start spinning faster but then go flatter as centrifugal overcomes aero. Point here is, though, that the movement is accelerating in its nature. Well, your standard recovery is the same as the reversed recovery if you’re inverted. It may pull the T-37 from inverted to erect and it may do so with your light aerobatics, but that doesn’t mean that this is what that step does.”
(26) additional comment against the BSWorks-5 video’s understanding of spins which Pages thought too much cumulative in the preceding endnote: “Have issue with the control surfaces work normally because of gyroscopic forces.” This is plain wrong. Control surfaces need to deflect air to create forces, were they acting directly on gyros, they’d not need interaction with the air. Yet they won’t work in vacuum and have less effect higher in thinner air. Gyro impacts are second order. The control surfaces still work aerodynamically. But, as you did elsewhere, you need to account for the relative wind. That much more aircraft bottom oriented axis of arrival matters to why the control surfaces work normally. Consider a swimmer cupping his or her hand. This grabs a greater volume of water. The same is true for the air of the down deflected aileron or elevator. Alternately, you could consider this to be a parachute of greater volume. The up deflected aileron or elevator lets more air by and is akin to a narrower parachute presenting less downward surface area. These cause shifts in the aero forces which in turn create moments upon the “flywheels” generating the gyro effects. But your control surfaces act normal sensing because of aero, it is just a different piece of aero than we typically consider.” In comment noted in endnote 25, I should have said spin axis moves forward of cg, not necessarily forward of plane. It is aligned with vertical axis through cg if flat, would be aligned in hypothetical pure roll ‘spin’ with longitudinal axis if pure steep, but between the two slides progressively forward till it would come back moving to the non-existent hypothetical pure roll ‘spin’ and may but need not be forward of aircraft. Such puts the plane into the helix & the helix helps explain how we can have high AOA in steep spins; consider we essentially throw our butts forward in our rotation.
(27) pg 523, Flying Qualities and Flight Testing of the Airplane — Darrol Stinton
(28) pgs 523 & 526, Flying Qualities and Flight Testing of the Airplane — Darrol Stinton
(29) image of slide 34 USAF TPS 2007 Spin Recoveries presentation
(30) pg 3–10, 2012 USN T-45C OCF FTI
(31) pg 2–21, 2012 USN T-45C OCF FTI
(32) pgs 64 & 88, Stalls, Spins, and Safety — Sammy Mason
(33) image of slide 29 USAF TPS 2007 Spin Recoveries presentation
(34) image of slide 5 USAF TPS 2007 Spin Recoveries presentation
(35) pg 109, Stalls, Spins, and Safety — Sammy Mason
(36) pg 125, Stalls, Spins, and Safety — Sammy Mason
(37) pg 544, Flying Qualities and Flight Testing of the Airplane — Darrol Stinton
(38) pg 128, Stalls, Spins, and Safety — Sammy Mason
(39) image of slide 43 with speaker notes, USAF TPS 2007 Spin Recoveries presentation
(40) pg 3, 1936 NACA Technical Note 555 — W.H. McAvoy https://ntrs.nasa.gov/api/citations/19930081370/downloads/19930081370.pdf
(41) pg 4, 1936 NACA Technical Note 555 — W.H. McAvoy
(42) https://www.youtube.com/watch?v=yyR9vCVWhZg
(43) pgs 398 & 400, Flying Qualities and Flight Testing of the Airplane — Darrol Stinton; image pg 401
(44) pg 13, 1957 NACA Status of Spin Research For Recent Airplane Designs — Anshal Niehouse, Walter Kliner, Stanley Scher https://ntrs.nasa.gov/api/citations/19930089716/downloads/19930089716.pdf
(45) pg 506, Flying Qualities and Flight Testing of the Airplane — Darrol Stinton
(46) pg 86, Stalls, Spins, and Safety — Sammy Mason
(47) https://www.youtube.com/watch?v=eTl8bK-saPA
(48) Catherine gets this stuff; she needed to simplify for a general audience with little subject experience, https://aopa.org/news-and-media/all-news/2020/march/pilot/flying-life-technedure “Catherine informed me that, yes, that was a general acronym, but that it wouldn’t cover the nuances of spin recovery for all aircraft types. In some aircraft, such as a Piper Tomahawk, the published recovery procedure is to apply full opposite rudder to the stop, then push the control wheel fully forward, and then check that the throttle is closed. For a Cessna 152, however, the throttle is the first thing to check as reducing airflow over the elevator will reduce the angle of attack and be the most effective way to stop the spin in that particular aircraft. There are even some military aircraft that require aileron into the direction of the spin. News to me! I believed my PARE acronym always was the way to recover from a spin, and I had passed on that bit of incomplete knowledge to many students through the years.” (Though you’ll be ok pulling power first in the Tomahawk. Their more ‘unique’ aspect is the pause after rudder then deliberate column forward more akin to the British method from which the Tomahawk will recover just fine. PARE will also work, it is just not what the manual says. After all, in PARE, elevator still follows rudder. Catherine is correct that PARE will not work in all airplanes, however.) Why is Catherine good while Juan and Scott are not? Because she talks in conditionals while they tend to talk in absolutes.
(49) https://www.segelflug.ch/wp-content/uploads/2018/10/ASK21_Spin_Test.pdf
(50) pg 19, July 1989 Air Force Flight Test Center Final Report, Schleicher ASK-21 Glider (TG-9) Stall and Spin Evaluation — Doyle B. Jansen & Charles J. Precourt, AFFTC-TR-89–27
(51) pgs 93 & 102, Appendix H, AFFTC ASK-21 Stall and Spin Evaluation
(52) pgs 88 & 89, Stalls, Spins, and Safety — Sammy Mason
(53) pg 16, Dec 1971 NASA Technical Note D-6575 — James S. Bowman Jr, https://ntrs.nasa.gov/api/citations/19720005341/downloads/19720005341.pdf
(54) seems in at least one case, they did test a light twin. Yet the POH of the BE-76 Beech Duchess still only gives the ‘spin prevent’ steps while stating they had not done spin testing as part of certification. Would be nice if they were less worried about liability instead giving both the prevent and discussion from their test efforts not associated with certification so as to describe the spin characteristics and recovery for us: Twin Spin Testing Beechcraft Twin Video https://www.youtube.com/watch?v=wzM1VoAnwvc — knowing such might even lead you to prefer the Duchess for multi-engine training as while Vmc Demo will still be a risk, it is a more mitigated risk than in other platforms.
(55) pg 83, Stalls, Spins, and Safety — Sammy Mason
(56) pg 124, Stalls, Spins, and Safety — Sammy Mason
(57) pgs 116–117, Stalls, Spins, and Safety — Sammy Mason
(58) pg 1–11, 2012 USN T-6A/B OCF FTI