Keeping the wings level in a stall
 

I would like to receive experiences and opinions on the techniques taught and used for keeping the wings level during deliberate approach to a stall.

It has become apparent to me that the technique I have been taught, and practiced in many types over many years, is not universal. I am required from time to time to stall different aircraft as a part of evaluation of their flight characteristics. When flying alone, I employ the technique I have been taught (it's kinda instinct now), and which has been reinforced to me by those who oversee what I do. However, on occasion, I am conducting stalls in the company of other pilots, and even with preflight briefing, find that some of those pilots use techniques which differ considerably from mine.

Though the wisdom of the experienced pilots is valuable, the new pilots here are equally important, as they have just received the training, and perhaps do not yet have "habits" overlying that training yet!

Tell me how you keep the wings level, and why you do it that way...

Thanks, Jim

I keep the ball in the middle with rudder input and gently release rudder pressure as I reapply power

Use rudder to keep wings level/ball centered, never use aileron near or at the stall (unless you want to spin :uhoh:)

As said, gentle movements with rudder to keep the ball centered (this is actually the most important thing), ailerons in neutral (try not doing so in proper stall training aircraft, e.g. Tomahawk and you'll see why), the elevator is in a position for required angle of attack.

As for recovery, full throttle isn't optional, speed of releasing (and pusihing) the stick varies from instructor to instructor. On proper aircraft it is neccessary to push the stick forward a little in order to brake stall, but for C172-like it is enough to just release stick and the nose drops. In my opinion, nose should be lowered (and thus reduced AoA) as quick as possible, disregarding of passengers feelings (as one of my instructors said, stall recovery should be slow and smooth), since stall isn't really state of aircraft I'd like to be in with aircraft full of passengers.

The technique employed can be type-specific. For example, one (modern) type I fly has effective ailerons right through into the stick hard back, fully stalled state, but be wary of any rudder input. So, horses for courses.

Rudder.. though mishandling that may (a/c dependent) tend to pitch one into rotation, as much as aileron missuse may on others. Also, in my experience, once fully stalled, it can be remarkably difficult to enter the spin, even when you're trying!

As for breaking the stall, I'm firmly of the stick and more stick principle. Primacy perhaps as I learned on gliders, but the 'minimum height loss' - pitch level and apply power technique really irks me. Fly the wing; it can be stalled or unstalled at any attitude. That said, there is no need to be violent, just positive.

My 2ps worth...

Why do you want to? Is that really important / a priority?

2 priorities are:
1. Unstall the wings - only now do you have "control" of the aircraft.
2. Minimum Ht loss in the whole stall / unstall / climb cycle.

1 is achieved by reducing the AoA. Not "pushing" the stick, but relaxing back pressure / easing it (centrally) forward until the wing is unstalled and no more. The moment the wing is unstalled, you have ailerons to level the wings.

2 is achieved by:
a. Only lowering the AoA until the wing is unstalled, and not "shoving" the stick forward.
b. Power
c. A prompt ease out of "dive" and into a climbing attitude (without further stalling the aircraft).

Clearly some teaching / types concentrate on the rudder - fine, if that is on good authority (in authorised manuals) - but I cannot recall seeing it on types I fly.

The aim is a simple repeatable drill that is being practiced for a stressful / hazardous situation. Rudder in the stall might work for a TP, but can also provoke a spin if used incorrectly, so IMHO not a good tactic. A prompt reduction in AoA as priority #1 will minimise any wing drop...

As a (Q)FI on various types (SEP, Jet) the usual technique errors I observe / debrief / re-practice on are excessive stick forward / nose down attitude, and then far from optimal transition back to a climb e.g. lose more height / eventually climbing 10K+ above what could be achieved.

NoD

Why would you want to? Simple, to execute a falling leaf, in a Tiger anyway. Great fun and a great challenge.

Perfect, more contributions please. Keeping the wings level to the greatset extent possible through the stall is the objective....

I was always taught to keep the aircraft balanced but DON'T try to pick up a wing in the stall using rudder. Simply unstall and then worry about wings level.

It depends upon what you are trying to achieve. Firstly, you initially talk about the 'deliberate approach' to the stall so, in the early stages (for argument's sake pre-stall warner), I would continue with the ailerons. At the later stages, the key bit is to keep the controls centralised and remain balanced (ball). As this is an exercise, if you have an unintentional roll at this point onwards, I would recover pre-stall. Conversely, if you are attempting to generate/demonstrate a wing drop then continue as suggested above.

In you last post you talk about the objective being to keep he wings level throughout the stall. Can I presume this means you stall the aircraft and are not seeking immediate recovery? Instinct tells me that, in general terms, pilots shouldn't be fiddling with any of the aircraft controls (aileron/rudder) during a stall unless they are trying to achieve something other than recovery. Under such circumstances, the actions are very dependent upon aircraft type.

In the PA38 Terrorhawk I was taught to use the rudder to keep the wings level; apparently the ailerons were ineffective during a proper stall and could cause a spin...

Otherwise, I use the ailerons to keep the wings level, and use the rudder to keep the ball in the middle because yaw during a stall is potentially bad news.

Certainly on the TB20 this is fine; it does everything it says on the tin, with no suprises, ever.

Recovery is a smooth re-application of power, back to the cruise setting, together with moving the yoke forward, so one recovers to cruise speed without any real loss of altitude (under 100ft). In the FAA IR they get you to do this partial panel ;)

Obviously an emergency recovery would use a bit more power than that :) but if you quickly whack the throttle fully forward you suddenly have a lot of yaw to deal with; don't try this in a TBM, evidently...

This is also fine in my Rallye where I can use the ailerons even when completely stalled and falling with style. However the input of aileron on some types can increase the stall to one wing (the down aileron side) due to the increased AoA on that wing and so likely if you tried to pick up a wing this way in some aeroplanes the aircraft would just roll more towards the wing-down side and you may well depart into a spin. I know you know this but I'm pointing it out for the benefit of the audience :}

I remember being checked out in the C-172 after getting my license in the C-150.

The instructor stalled it; then with the stick all the way back, flopped the ailerons back and forth from stop to stop -- wings stayed level.

The glider checkout instructors seem to get a bit miffed when I use rudder to stay coordinated approaching the stall:confused:

The rudder should be used to STOP the yaw this should prevent wing drop.

Rudder must NOT be used to "pick up" a wing only to prevent further wing drop.

Stall + Yaw= Spin!

The only way to create a rolling moment with the rudder is to induce a sideslip. Rhetorical question: Would we rather stall in coordinated or uncoordinated flight?

The scenario: Reduce the airspeed, add back pressure, ailerons neutral, apply rudder pressure. I just described entry into a spin. To me it sounds a bit peculiar to have this scenario described as the proper way to prevent a spin.

What I was taught, and use, is to apply slight, smooth aileron inputs to keep wings level, using rudder for coordination. At the stall, ailerons neutral and use rudder to counter any yawing tendencies. If a wing drops at the stall, e.g. if one uses too much aileron or flies a plane that does have a tendency to stall over a wing, then that is just the way it is; it will be a stall with a wing drop. Recovery from a stall with a wing drop is something we all learned during training, right...?

Even if one uses too much aileron and provokes a wing drop, a spin should not result provided one uses rudder to counter adverse yaw and maintain yaw control.

Apart from my training, both standard references I use for basic flight procedures and techniques (the FAA's Airplane Flying Handbook and John S. Denker's See How It Flies) suggest coordinated use of aileron and rudder to maintain wings level during the approach to stall.

Nevertheless, intrigued by what I have been reading here in the last few years, I decided to go up and try the "keep wings level using rudder with ailerons neutral" approach to see for myself. I used a C172. After some rather spectacular departures for a C172 :E I decided it was just as bad an idea as I thought before I tried it.

The problem with that method, IMV, is what happens if it fails. If you manage to keep the wings level with small inputs, i.e., do not have to try and catch a wing that shows a tendency to drop with either rudder or aileron, both methods work. But if you fail, the coordinated aileron method results in a rather benign wing drop, whereas the rudder only method results in an incipient spin.

I would suggest you should approach this question from a different direction.
Assuming we are not talking about aerobatic training, I think it is imperative that one does not think about the stall entry as a stand alone exercise. There is no reason the aircraft should stall at altitude. If the aircraft has stalled than the guy/gal at the controls has seriously screwed up and if close to the ground this is a full blooded emeregency. Therefore it is vital that the instinctive reaction of the pilot to a stall is to lower the nose to unstall the wing and to simulataneously use all necessary rudder to stop the aircraft from yawing thus making it impossible for it to enter a spin. In training the worse the stall entry the better as if the student can reliably recover from a banked yawed entry than they will be well prepared if they ever get cought for real. Finally the best recovery from a stall is to not stall in the first place therefore training to recognize the condition which indicate a potential stall is possible is just, if not more, important than stall recovery training. Stall entry training is simply a necessary precursor to learn the real lesson, stall avoidance and recovery, and should never IMO be presented as an exercise in itself.

Why has no one specified what type of stall they are doing, power on, off, accelerated?
In most single trainers, doing a power on departure stall, you will most certainly drop a wing at the break of the stall if you kept the ball centered and the wings level during entry, does anybody know why.

Applying that principle to engine failures would have eliminated one of the longest current threads on pprune!

barrow, consider the relative airflow generated by the propeller in a power-on stall.

Torque, slipstream and P Factor all spring to mind.

The point of my question was the technique used to fly the aircraft to demonstrate compliance with the certification requirement (either power on or off) that it be possible approaching and throughout the stall to maintain the wings within 15 degrees of level, with normal use of the controls.

We do it at altitude, 'cause it's safer up there. I agree that there is no good reason to do this in "regular" flying, other than training, or flight test.

CAR 3.120 (c):

....and not more than 15 degrees roll or yaw shall occur when controls are not used for 1 second after pitch starts and are used thereafter only in a normal manner.

During flight testing I have done with other pilots in the past, who have suddenly decided to "help" me with the stall recovery during such demonstrations, I have seen the ball hard over (where I had been carefully keeping it centered) when the other pilot applied a boot full of rudder. This un-nerves me, as being rather close to a speed at which a stall or spin could be expected, the last thing I feel comfortable with is seeing the ball hard over. To me, that is spin entry territory, and a spin is not the objective.

I have reviewed the 22 different flight manuals in my library today. None mention the use of rudder during a stall. A few mention the normal use of ailerons. Many just refer to recovering (with no stated technique), and a couple do not mention stalls at all. The FAA flight test giude requires the "normal" use of controls. I hold the opinion that the use of rudder (without accompanying aileron) to un co-ordinate level flight at any speed would not be normal use of controls.

So with the certification requirements requiring a demonstration of wings level with "normal" use of the controls, what would be the basis of training not to use ailerons during the approach (be it deliberate, or suddenly recognized) to a stall? Is the rudder only technique written somewhere?

Given the various types of stall available, from your basic straight level through to fully inverted stalls along with steep turns keeping the wings level is irrelevant. After the stick back pressure has been released part of the recovery is to get your wings level as the angle of attack is relaxes then power on and climb. There was an excellent article in Australian Flying about 5 years ago on the "stall stick theory".

For those that really want to stall grab yourself an instructor climb up to 4500agl slow down to just above stall, reef the stick back as far as your can cleanly then kick in a full boot of rudder and enjoy the ride, i repeat dont try this without an instructor.

Yes, but the stall will occur first and provided you recover from the stall the spin will not develop.
In a perfect aeroplane with perfect rigging, flown perfectly in balance with no propellor effects, both wings will stall at the same time and a wing would not drop.
As these conditions will not exist, if you want the aircraft to stall both wings at the same time, I would suggest that the rudder is used to prevent or induce a yaw to try to cause each wing to stall at the same time. To achieve this may well involve the balance ball not being centered but the result is each wing will stall at the same point due to all the variables being equal.

Nothing to do with a wing drop in a stall.

Then why does everyone keep the ball centered to the stall, then wonder why the wing drops!

bingofuel gives the right answer, You want to stall and keep the wings level throughout recovery, don't center the ball, less right rudder and a slip is needed.
Ailerons are used for roll control and rudder for yaw. Using rudder alone to keep wings level during entry is poor form.

I'd suggest that's a massive oversimplification. Take a look at secondary effects of controls, and (I don't know if this is taught powered - it wasn't when I converted to spamcans), how the primary and secondary effects change (actually swap over) with speed.

Arguably, a 'proper' spin doesn't exist until something like a whole turn or more - however, that's probably moot to this discussion, and a little purist.. but I do wonder how many are posting 'this will spin' know so from practical experience, or are just repeating - sometimes, aeroplanes can be surprising things:

Take a decathlon, establish a full sideslip, either way, rudder on stop, and opposite aileron/wing down to hold a straight line. Slow up until it stops flying, and the result is?

A (repeatable) clean, straight ahead stall with no wing drop? Who'd guess?
I wouldn't extrapolate to all a/c, particularly as the decathlon has a convenient straight, high wing with next to no dihedral. No matter the angle the wing is travelling, all parts are meeting the air at the same angle and speed. Why would one end stall first?

Personally I'm inclined to believe any static yaw is distinctly secondary to *rate* of yaw in determining wing drops. Obviously sweep and dihedral make yaw matter more.

BTW, those 'perfect' conditions do exist, It's called a glider :E

You are closer to a spin if you stall with wings level and a side slip than if you stall with a wing drop and no side slip.

Spin requires stall, departure (i.e. none or negative roll damping, i.e. uncontrollable roll) and yaw (e.g. side slip). If you set up a side slip before the stall in a misguided attempt at preventing a wing drop you are already one step closer to the spin, especially keeping in mind that a side slip will lead to the wing drop. The departure/wing drop also leads to a yaw eventually, but that yaw tends some time to build up.

By recovering from the stall after the wing drop but before the yaw builds up you will prevent the incipient spin from developing. Doing it the other way around, recovering after the yaw but before the wing drop, will not be likely to work.

In my admittedly limited experience, yawing/side slipping at the stall will always lead to a wing drop before the stall recovery, but a wing drop at the stall will not lead to significant yaw or an incipient spin. I know (as in have been told that) some planes will always produce a yaw after the wing drop before stall recovery is complete, but I do not believe there are many planes where a wing drop produces yaw but where a yaw does not produce a wing drop. Consequently, preventing the yaw is always the number one priority to avoid an incipient spin, whereas preventing a wing drop is the second priority.

For Pilot DAR's test flights, I don't suppose the testing criteria will be met if he stalls with wings level and a side slip... :8

As for why the wing drops: Preventing a wing drop does not necessarily require that both wings stall at exactly the same time! Some roll damping can still remain even if pitch stability is lost. This is because roll damping is primarily provided by the outboard parts of the wing, whereas pitch stability and vertical damping is provided by the entire wing. If the inner wings are sufficiently stalled for vertical damping to be lost (i.e. the aircraft settles straight ahead) or pitch stability to be lost (i.e. the nose drops) while the outer wings are sufficiently unstalled for roll damping to remain, the aircraft will not drop a wing. That is how modern aircraft are designed.

If preventing a wing drop was dependent on both wings stalling simultaneously, it would be a rare exception that an aircraft would stall without a wing drop. That is not how it works at all. It is the roll damping that prevents the wing drop.

In other words, maintaining aileron effectiveness during a stall is a secondary reason why we always want the inner wing to stall first. The primary reason is to maintain roll damping during the stall. Preventing a wing drop by having ailerons effective but no roll damping, thus relying on the pilot to balance the plane and actively prevent a wing drop, would be very hard and rarely successful. Preventing a wing drop by having remaining roll damping is automatic, the pilot doesn't have to do anything. Roll damping during a stall is the most important thing; aileron effectiveness is secondary.

And contrary to what a previous poster said, the airflow generated by the propeller has everything to do with why a wing can tend to drop at the stall; at least it is one of the most important factors. For example, the helical prop wash hits the two wings at slightly different angles of attack, tending to cause them to not stall at the same time near the roots. If sufficient roll damping remains, this will still not result in a wing drop, but if there is insufficient roll damping then this assymetrical slip stream effect is one of the reasons why a wing will drop.

On the relation between roll damping and use of aileron: Roll damping is what stops a roll if it develops; aileron control is used a return the wings to level after the roll damping has damped out any roll disturbance. If roll damping is so low that the pilot actually has to balance the wings using continous aileron input, the pilot will most likely not succeed for very long and this will cause a wing drop. If the pilot tries to achieve the same thing with rudder, the risk of over controlling is much greater (since the rolling moment is a secondary effect to rudder input, after a yaw has already built up), and the result if one fails are much more dramatic.

Again: Attempting but failing to keep the wings level with rudder results in all three pro-spin factors to be present; stall, yaw and roll. Attempting but failing to keep the wings level with aileron only results in two factors to be present: stall and roll. In the latter case, recovery can be made before the yaw builds up.

Edit: Forgot to point out that when I mention "aileron input" I of course mean coordinated aileron input! As always (except possibly in cruise in some aircraft...) when the aileron moves the rudder should move as well. But this rudder movement is not intended to cause a yaw but to prevent a yaw. Getting that wrong and using only uncoordinated aileron will of course cause a yaw and a side slip and that will mess up everything. But this coordinated use of rudder to prevent the adverse yaw due to the aileron input is completely different from the uncoordinated use of rudder to cause a side slip with a secondary rolling moment.

On flight tests you ask the candidate to demonstrate a stall. Invariably they close the throttle and make no attemt to compensate for the yaw caused by reducing the slipstream. Invariably, they apply some aileron to keep the wings level thus, if you hold the aircraft in the stall a wing is likely to drop.

As you approach the stall keep straight with rudder, never mind the ball, just look out of the window and pick a point in the distance. If you prevent yaw the wings will remain level until the aircraft stalls.

What has "primary and secondary effects" got to do with stall entry? Nothing! and to mention it as a factor is just plain silly.

bjornhall, 12 paragraphs of utter nonsense. the only part of your post that made sense was "In my admittedly limited experience"

Single engine trainers, (US) are built with a off center thrust line, higher angle of incidence on the left wing and a canted vertical stab to offset the LTT in cruise. When the AC is stalled with a centered ball, both wings have an unequal AOA at the moment of stall, this leads to wing drop.

barrow, just because you haven't understood something doesn't make it less true...

Go up in an airplane and do some experimentation on your own then. You should be able to discover that what you just said is not relevant to what actually happens.

There will be all sorts of asymmetries in any stall situation aside from the design asymmetries you mention; a small slip angle, a small gust at the wrong time, slightly unequal fuel or payload distribution, and so on. If such small effects were enough to cause a wing drop it would be a very rare event to see a straight ahead stall. But what happens in real life is that modern aircraft with sufficient washout etc tend to stall straight ahead every single time! The reason is that the roll damping catches the roll before it has time to develop, and only a very slight residual bank remains to recover. Therefore, contrary to what you say, the design asymmetries do not cause a wing drop, even if the ball is not held slightly off-center to compensate for the asymmetric thrust.

If you don't have sufficient roll damping for whatever of many possible reasons, you do get a wing drop if you have any asymmetry in the situation. But since the design asymmetries included to counter the LTT in cruise are just one example of such asymmetries, and there could very well be some other asymmetry at play (see examples above), trying to fly with exactly the right amount of off-center ball to counter the design assymetries would be a waste of time IMV. Indeed, if you would try that and get wrong the resulting side slip would cause a rolling moment due to slip roll coupling that could very well be far bigger than the design asymmetry you tried to compensate for in the first place.

The asymmetries you mention really are extraordinarily small. To provide slip free operation with one engine out in a twin requires a bank of a couple degrees or so. That is with the operating engine mounted all the way on a wing; how many thousands of a ball width will be required to offset the off center thrust in a light single? Not many!

BTW, you should probably be aware that trying to emphasize your points by insulting those you discuss with just might make you come across as a somewhat mentally challenged individual... Just thought you might know, to make a better first impression next time.

I'm not experienced at experimenting with the stall, but I had a trial flight in a pitts a couple of years ago where it was demonstrated how to descend in the stalled condition by keeping wings level with the rudder alone (stick fully back).
A boot full of rudder in this condition however would result in a flick roll.
At least, that's how I remember it.
I haven't tried it in a cirrus though!

I shall also assume good intent:

At normal flying speed, ailerons produce mostly roll, and some adverse yaw (secondary effect). The rudder produces yaw, and eventually roll (secondary effect). I'm sure you know that.

Near the stall, (i.e. at high angles of attack) the ailerons will produce a far more pronounced yaw, and reduced roll response. The rudder will produce less yaw, and a more pronounced rolling tendency. In some types you will find that at some speed near the stall, the secondary effect becomes the dominant. Not to mention the already made point that in many types, aileron use at the stall will create a wing drop in precisely the opposite direction to the roll 'commanded' by the ailerons. That may have nothing to do with stall entries, but it has everything to do with your assertion that:
Bjorn,

What you say makes sense to me in the main, but I didn't like the idea of the roll damping 'catching' the roll. Unless I'm mistaken, damping is just a resistance to movement (in this case the roll), an inertia if you like. It will not actively restore the wings level, rather make the aircraft less susceptible to the 'fine detail' level of the wings stalling *exactly* at the same time? I'd still suggest they need to stall substantially together.

I'm also interested in the definition of slip and particularly yaw. Does an aircraft in a stable, straight line sideslip have yaw? It definitely has slip, and I know the fuselage is not pointing in the direction of travel, but I'm inclined to suggest it doesn't have yaw, and if it does, it doesn't have a rate of yaw - all parts of the wing move at the same speed, and in terms of direction and space, it is not moving around the yaw axis. In a sense it's no different to an oblique wing NASA AD-1 - Wikipedia, the free encyclopedia flying in a perfectly straight line.

At one point in my aero's training I didn't like spinning the Decathlon.... so my instructor and I approached it by getting me to hold in a fully developed stall, holding wings level with rudder (stick fully back), and not recovering until I'd descended 1500 feet.

We then clambered back up and tried again but allowing an incipient spin by letting the wing go....

I got there in the end, and managed my 3 turn spin and recovery...

bjornhall, Are you kidding me pal? Again 6 paragraphs of utter nonsense from you in an attempt to bolster a completely flawed conclusion.

Which types? please specify.

So, if there is merit to using the rudder (disproportionately to the ailerons)to control roll at low speed, and during the approach to a stall, why can I not find a flight manual which says this is the way to do it? I have to presume that if the flight manual says to use "normal" control inputs, or is silent on technique, that "normal" use of ailerons for roll, and rudder for yaw, is intended. Are there authoritative publications out there which describe these "rudder priority" techniques, or are they only training "hand me downs"? If so, from where?

I did about 30 minutes of stalls in a Grand Caravan today. All unacellerated, but all power and flap settings. When high power stalls were approached, a boot full of rudder was needed to overcome the torque, but the aircraft otherwise flew level and ball in the middle. When it broke in the stall, it went wings level, with only very slight aileron inputs to keep it so (the 208B POH is silent on specific stall handling technique).

So where do I find a document which describes the primary use of the rudder near the stall?

Keep the ball centered with the rudder. Don't pick up a wing with ailerons when stalled.

Proper stall recovery:

Push to unstall wing

Center the skid ball with rudder

Apply power and resume flight.

Applying power before unstalling the wing will cause a lot of torque roll, in some planes you will go inverted, Pitts, Midget Mustangs ect.
If you try to pick up the wing with aileron, when stalled, you will further stall the wing with the downward deflecting aileron. Remember your spin recovery practice? Trying to use aileron to roll out of the spin accelerates the rate of spin. The reason being that you are putting that wing with the downward deflecting aileron deeper into a stall.

It makes logical and aerodynamic sense NOT to use ailerons very close to the stall...We all know that ailerons change the AoA of the wings and so if you are happily flying at max AoA (say 17 degrees) on both wings, nicely balanced and you put an aileron input to the right, then the left wing AoA could exceed max AoA and the right wing could be below max AoA. In this instance the left wing will stall and the right wing won't. Using aileron to further try to pick up the left wing will exacerbate the situation and the roll will continue.

Doesn't happen very dramatically on a training aeroplane where the insides of the wings tend to stall before the outsides and you maintain aileron authority even into the stall but I imagine on something like an Extra with an asymetric wing the effect would be quite pronounced. Sometimes it helps get aeroplanes into a spin - one FI I knew used to spin the C150 by applying full power, full left rudder and full right aileron to ensure it entered the spin.

We are into the fine tuning parts of the discussion here... ;) Everyone probably knows and agrees that a large aileron input close to the stall will often make that wing stall, and that an aileron in a stalled part of a wing will work in the reversed sense in that regard (aileron down on a wing will cause that wing to drop, not rise). I just do not agree that it is therefore a logical conclusion to not use ailerons close to the stall.

The argument against that conclusion is that in a wing with sufficient washout the part of the wing where the ailerons are mounted are not so close to the stall that they work in the reversed sense, even if the airplane as a whole is stalled (i.e. vertical damping and/or pitch stability is lost). Ailerons on modern certified aircraft (and on many others, although not all!) will work in the normal sense even in the stall, and certainly just before the stall. We just should not use sudden large aileron inputs very close to the stall, or try to pick up a wing that is already dropping using ailerons (if we already have uncontrollable roll due to loss of roll damping we will most likely have lost aileron control as well).

We don't have to use rudder and slip since we still have aileron control, and using ailerons is both a more efficient, faster and safer way of doing it IMV.

I think it is interesting to note that in the C172 the recommended control positions for spin entry is ailerons neutral or a slight aileron input into the spin (e.g. full right rudder and neutral or slightly right aileron into the right spin). Even in the spin entry the ailerons work in the normal sense.

Couldn't say how it works in an Extra.