Multi-Engine Training: Stall Recovery....

[Clodhopper]

flyin, I haven't flown one in years but if memory serves the flaps come up before the gear on a Seneca, or at least the Seneca I because the gear has neglible drag so it does no real harm to leave the wheels down.

When I learned it was Flap-Gear-Flap. Full flap produces far too much drag for its lift benefit. And leaving them out and then adding the extra drag of opening gear doors and moving the landing gear isnt wise. So you retract the first flap setting, reducing drag, and then go for gear, adding a bit of drag with the open doors, but not as much as both at once.

[justplanecrazy]

Back to the POH reasoning. I believe the reason they say flaps first and wait for a positive rate of climb before gear, is due to the situation your simulating. It's supposed to be training for a stall on approach, hence the gear down. If you stall fairly short final on approach you wouldn't want to recover with nose down, power, gear, flaps, climb. The result would be something like this:

You're on the approach and nearing minimums, the stall horn goes off and you realise you're now stalled so you recover
Yoke slightly forward (now really close to ground but wing no longer stalled)
Power added (airspeed increasing and starting to reverse the descent)
Gear up (drag reduced and descent just about stopped)
Plane is now planted gear up, flaps down on the threshold of the runway.
Welcome the passengers to their destination and call a mule

I think you'll recover faster if you do gear first as flaps produce lift and lower the stall speed and gear doesn't. The reason you don't is the only place you'll have the gear down is beacon final and you'll be close enough to the ground that a stall presents a good possibility of hitting the ground. When you're in the muck, you don't want to pull that gear until you have a positive rate of climb and know that you are no longer descending closer to the ground just underneath you. If you stall at minimums, there's a good chance you'll recover from the stall but not have enough time to start climbing before you touch down. If you have the gear down, and you're lucky, you may just get a hard landing.

If you bump the flaps to ten, you won't lose all the lift produced by the flaps but you'll lose most of the drag and only increase your stall speed to the 10 degree flap speed, not the clean configuration. The result is a faster turn around from descent to climb. Then when its positive and you're no longer in danger of hitting the ground, pull the gear to get rid of all that drag with no lift and finally go to 0 flap. Different types, different recoveries, this is just a generalization for most twin trainers.

Basically its switching from your planes best approach config, to the best short field departure config. Then when you have a positive rate, just like on departure, you pull the gear and continue as you would with a short field takeoff procedure.

[Jimmy Mack]

stall horn goes off and you realise you're now stalled

shouldn't the stall horn go off prior to the stall? I believe it is a stall warning device, not a stall indicator. In which case, the wing is not stalled and thus you don't have to check forward and lose valuable altitude.

JM

[co-joe]

Why are the flaps coming up before the gear?
Flaps improve the wing's aerodynamic performance at lower airspeeds.
So, raising the flaps at a slow(er) speed with the gear down doesn't make much sense. Is this a POH thing?

The answer seems to depend on what aircraft you fly. The Beech AFM mentions mothing about how to recover from stalls so it's up to interpretation. If the POH/AFM for your particular aircraft says what order to follow then I'd follow it.

But here's our SOP's for the stall in the landing config.

1. Max Power
2. Flaps approach
3. Positive Rate then gear up
4. IAS through blue line flaps up, climb power, after takeof checks.

It makes sense not to check forward first since in the landing configuration you are probably about to land right? So breaking the stall will still result in ground contact. Max power first since the induced lift of the prop wash will in effect lower the AOA as well as provide induced lift thus lowering the stall speed.

Blue line is irrelevant here because both engines are working and we need to keep from hitting the gropund at all costs. If max power doesn work, the next step is Firewall Power.

[justplanecrazy]

stall horn goes off and you realise you're now stalled

shouldn't the stall horn go off prior to the stall? I believe it is a stall warning device, not a stall indicator. In which case, the wing is not stalled and thus you don't have to check forward and lose valuable altitude.

JM

Yes, yes it will and that is why I said you hear the stall horn go off and then you realise that you're actually stalled. I thought it was painfully obvious that we were talking about stall recoveries not what to do if you hear the stall horn.

Gee co-joe, the BE20 doesn't have a stall recovery in the POH???? If you bought the plane privately and didn't have SOP's, you'd be screwed!!! Do I do 10 or 20 flaps!!!! Damn it, it doesn't say I'm going to exceed the design limitations!!!!! Damn you BEECH AHHHHHHHHHH!!!!!!!! We're all going to die!!!!!!!!!!



Sorry couldn't resist, after posting the BS that you have under the flaps post, you better expect some jokes heading your way for a long time.

As far as your breaking the stall comment, isn't lowering the stall speed and angle of attack, breaking the stall. I'm not saying dive for the ground, just allow the angle of attack to be reduced by moving the yoke slightly forward and add power. In a small light twin, you'll lose a lot less altitude if you don't hold the same attitude but allow the nose to pitch down a little as you add throttle. But what do I know, I'm not quoting out of a POH or anything.

[co-joe]

Yeah I saw that one coming.

The point is that if the AFM doesn't say what to do, that's where SOP's come in. If there are none as in your type of operation then try common sense, (which you seems short on). But hey, if you tend towards being a cowboy then YeeeHaw! Do whatever works eh!

In a beech you don't need to check forward to break the stall, power and drag do it for you.

In a seneca you may need to check forward a little, but I'll defer to the experts on that one. I only have one seneca flight under my belt.

[Doc]

co-joe has it about right. Just go max power to arrest descent, flaps to approach, pos rate, gear up...and off you go! Now, the whole exercise in a twin is to simulate the "approach to a stall", not a full aerodynamic stall. Usually we do this in a turn, to simulate a circling approach (the only time you will be flying around at a low airpeed with everything hanging, unless, of course, you're a complete idiot!) when the PNF dosen't keep an eye on the airspeed.

[Clodhopper]

I think you'll recover faster if you do gear first as flaps produce lift and lower the stall speed and gear doesn't. The reason you don't is the only place you'll have the gear down is beacon final and you'll be close enough to the ground that a stall presents a good possibility of hitting the ground. When you're in the muck, you don't want to pull that gear until you have a positive rate of climb and know that you are no longer descending closer to the ground just underneath you. If you stall at minimums, there's a good chance you'll recover from the stall but not have enough time to start climbing before you touch down. If you have the gear down, and you're lucky, you may just get a hard landing.

Also a good point, I'll admit, I had my mind in the windshear mode, where you leave the gear down and get the power on. My mistake. You're right about leaving the gear down as a safety factor for ground contact.

[Rowdy]

Here's what tickled me.. and I'm not an instructor nor a high time twin pilot, but.. with the terminology check forward.. I believe that might confuse some of the other young, just starting out CPL guys..

Correct me if I'm wrong.. but wouldn't you want to just check level? check to far forward near the flair or on the approach and that could be kinda rough on the old gear hanging down right?

[goldeneagle]

In a beech you don't need to check forward to break the stall, power and drag do it for you.

Actually, this is not true. There is one, and ONLY ONE thing that will break the stall, you must reduce the angle of attack. There are a number of ways to reduce the angle of attack, some generic, some airplane specific.

a) poke the nose down a couple of degrees from it's current attitude. Generic solution, works in EVERY airplane with fixed wing configuration. Breaks the stall, doesn't stop the descent.

b) retract flap. Works in most airplanes with conventional flaps, it's effectively raising the trailing edge, so, reducing the angle of attack. Side effect is reducing drag, allowing the aircraft to accelerate, which has a further side effect of reducing the apparent angle of attack. On airplanes where the flap retraction doesn't make a big difference to the overall angle of attack, then flaps are managed as drag devices, not lift devices, reference below. In either case, this can break the stall, does not check the descent.

c) Increase power. If the airplane has enough power (BE/20 does) this will cause a number of things to happen. First, it'll accelerate (which decreases the apparent angle of attack). Assuming you are in a nose high attitude, some of the thrust increase will resolve onto the lift vector. Do the full analysis, and you'll see, that negates a little of the weight vector, so it decreases the angle of descent, which lo and behold, decreases the angle of attack. This is only helpful if the nose angle is above level. If the nose is below level, and the goal is to minimize altitude loss, the increase of thrust can well work against you prior to getting the nose above level, the increase in thrust will add to the downward force vector, which is the sum of weight plus all downward aerodynamic forces.

Now, take a look at other drag producing appendages (speed brakes, landing gear, some flap configurations, etc). Tucking those out of the airflow will have little/no affect on the stall itself, but, lots of side effects on the 'big picture'. Typically the goal of stall exercises is to stop a descent, and convert it into a climb, which means configure the aircraft so the maximum amount of excess power is available to climb. Drag consumes excess power, so, you want to get rid of it. Gear is a special case, and one needs to understand the timeframes to manage gear correctly. If time is indeed very short, say 10 seconds to impact, and the gear retraction cycle is 15 seconds, then, if your gear door configuration increases the overall drag of the aircraft while the gear is in transit, you wanna leave it exactly where it is. You only want to start a gear retraction cycle if the time available allows for the full cycle to complete in the case of an airplane where the retraction cycle actually increases the drag on the plane. Most other drag producing appendages have no bad side effects during the retraction cycle, so, you want to get them moving to the zero drag configuration as quickly as possible.

And now, since you got me going on the subject, lets talk about SOP's for the exercise, where they come from, and how they are derived. Dunno if your operation covered this in the groundschool for the airplane or not, but, if not, the groundschool was deficient. The process works like this, first you assume the initial condition, configured for landing, with appendages hanging out, for this discussion, lets assume gear and full flap, max landing weight. I have no clue what the numbers are for your airplane, never flown it, but, i do know what they are for my airplane, so I'll quote those.

With the 421b, max landing weight, max power (firewall throttle) go around at sea level, clean configuration, you can expect to see 1400+ fpm on the vsi. With only one fan turning, it's going to be around 475. The POH quotes less than that for the single engine climb rate, but, read it carefully, the lesser number is max takeoff weight, go to the charts, and find it for max landing weight, it's around 475. Repeat the exercise for both fans turning if you question the 1400, again, it's landing weight numbers, not takeoff weight, with full power, not climb power. For each appendage hanging out, you will pay a penalty in drag, which translates to a climb ability reduction. Gear down, it's 400fpm, gear in transit it's 500fpm. Flaps full it's 700fpm. Propeller windmilling it's 400fpm. You wont find those numbers in the POH, but, you will learn them either by trial in the airplane, or spend the bucks on a trip to flitesafety, and they will spoon feed you the numbers, then prove it in the simulator.

So, now what do we expect from our airplane, landing configuration, max power at sea level. It's going to have a climb of 1400fpm, less 400 for the gear, another 700 for the flaps. With both fans ticking over cleanly, we can fly away and still show a climb with everything hanging out. The largest benefit comes from retracting flap, and we actually pay a small penalty for gear in transit. So, now it becomes fairly obvious what should be the standard procedure. Attitude to break the stall, power up with nose above level, flaps up (start them as soon as you can reach the switch if single pilot, call for the PNF to get them going if 2 pilots), confirm the positive rate of more than 100fpm, then select gear up after we have cleared any obstacles immediately ahead. Immediately is easy to define, takes 15 seconds for the gear to come up, at 2 miles a minute, that's a half mile, then add another 1/4 mile to capture the climb increase. Obstacles more than 3/4 ahead are insignificant, ignore them when deciding if you want the gear in transit. With obstacles beyond 3/4 mile ahead, you are better off with gear in transit immediately.

Now, analyze the worst case, hit the max power, and only one fan comes alive, the other stays windmilling. 475 fpm less 400 for the gear, and less another 700 for the flap, and another 400fpm for the windmilling prop. Not good, we are still descending at more than 1000 fpm. Get the flap on the way, and feather the bad one, our energy budget looks dicey, but, the plane can struggle to maintain level with the gear hanging out, it's going to start descending again if we get the gear in transit. So, now we have to look at our initial SOP and add some qualifiers. It's still power up, flaps up, but now we have to add a 'stop and think momentarily' item, check for yaw. This is a decision point, if there is no yaw, and the climb is confirmed, gear up and fly away normally. If there IS yaw, and the climb is not happening, it's time to get looking at guages, identify, and feather. There is a new decision tree that has to happen at this point, and it has to start with the question 'is it even possible to prevent impact?' You CANNOT be stopping to do the math above in this scenario, so, it must be done in advance, and, the SOP has to account for it. When you do all the math, you will discover, with the 421 at max landing weight, accounting for a very small amount of recognition time, and a small amount of reaction time, with less than 300 feet to spare, the single engine go from landing configuration just isn't going to happen, the physics say so, especially if we are trying to accelerate out of a stall condition. So, now we add another item in the SOP decision tree. If there is yaw when the throttles come up, and altitude is less than 300 feet with everything hanging out, use that 300 feet to choose the point of impact, you are landing short, lets land, not crash. There is one good engine that you can still use partial power on very safely, lets use that power and our brain to pick how and where we touch down. Forget the flaps, forget the feather, concentrate on a survivable landing. I emphasize this by now going to the SOP's for landing. At 300 feet, with everything hanging out, the PNF calls out '300 feet'. From that call to touchdown, the PF no longer considers go around assured, and in the event of a go, the onset of yaw means it aint gonna happen, no debate, just land. This is an airplane specific SOP that accounts for the limitations of the machine, all decisions are made in advance, no pilot thinking required, when the circumstance arrives, just execute. When the throttles come up, if the nose starts to turn, and you've heard the '300 feet' call, look out that window, choose your point of touchdown, and land.

Contrary to what you may believe co-jo, I do believe SOP's are a good thing. They allow me to sit back, thoroghly analyze a lot of situations, take the time to do the math, and figure out exactly what the airplane can, and cannot do. SOP's are written with the benefit of 'no pressure', time to think, a calculator, all of the performance charts, and the ability to call the engineers and ask dumb questions if necessary. Then we boil it all down into simple procedures, write them up in a list, and it goes onto the airplane. For time critical safety procedures, I require pilots commit them to memory, and know the flow of the decision tree. I also require them to know WHY the decision tree is structured as it is, so they can apply 'local knowledge' if necessary. The stall/go tree has to be adjusted if there's a half inch of ice on the plane, and a pilot that understands where the 300 call comes from, will probably look to the pnf on the way down and say 'call 500 feet today'.

<humor>
And that sir, is why I'm the chief pilot, and you are the co-jo
</humor>

[Hedley]

This may seem somewhat orthogonal to the discussion, but given
the frantic and panicked attitudes towards stalls (single or multi
engine) typically displayed by pilots, I would like to mention that
a stall does not cause an airplane to explode, or become uncontrollable.

I will frequently fly through a stall, both upright or inverted, in
close formation, quite happy at low altitudes. It's really not that
big a deal, just be aware of your Cl and Cd curves wrt AOA.

Take a look at the (negative) AOA I have on here, formating
on a tortoise-like buck fifty:

http://www.pittspecials.com/images/pg_inv.jpg

Quite frequently, during inverted formation with slow (100 mph)
aircraft, I will stall, and have one or both wings dropping gently
as they exceed the stalling AOA. It's really nothing to get
excited about. No need to panic, and get your hands all
a-blur in the cockpit.

So please, stop dissin' the stall - it's really not as bad as you
might think

If you want something honestly challenging, might I suggest
inverted spins below 1000 feet?

[justplanecrazy]

b) retract flap. Works in most airplanes with conventional flaps, it's effectively raising the trailing edge, so, reducing the angle of attack. Side effect is reducing drag, allowing the aircraft to accelerate, which has a further side effect of reducing the apparent angle of attack. On airplanes where the flap retraction doesn't make a big difference to the overall angle of attack, then flaps are managed as drag devices, not lift devices, reference below. In either case, this can break the stall, does not check the descent.

The stall speed in a 421 goes from around 95 to 85 when you have the flaps down, doesn't it? I'm not sure haven't flown one. When you're trying to break the stall, you have to lower the angle of attack, but unless its a power stall, you're airspeed is a good indicator at what you aoa is. If you raise the flaps prior to breaking the stall, you've just added another 10 kt's to your current speed that you need to increase. If the stall speed is 85kts, you'll probably actually experience the stall at 75kts or less unless your full gross forward c of g. Does it really make sense to reconfigure the plane so that instead of having to gain a couple of kts to have to break the stall, you now have to create 10 plus a couple more knots to break it? I don't know maybe your onto something here but unless you really hold a plane in a stall, the drag caused by the flaps won't be a greater negative then the positive produced by the reduced stall speed. A good way to test the theory is at altitude, idle, dirty, really hold the plane into a significant stall, then mark your altitude just before recovery and see how much you lose by trying to recover by just nosing forward without jumping the throttle. Then try it again by retracting the flaps and nosing forward to recover.

If time is indeed very short, say 10 seconds to impact, and the gear retraction cycle is 15 seconds, then, if your gear door configuration increases the overall drag of the aircraft while the gear is in transit, you wanna leave it exactly where it is. You only want to start a gear retraction cycle if the time available allows for the full cycle to complete in the case of an airplane where the retraction cycle actually increases the drag on the plane. Most other drag producing appendages have no bad side effects during the retraction cycle, so, you want to get them moving to the zero drag configuration as quickly as possible.

If you're taking more than 10 seconds to recover from a stall, you're in trouble. If you start thinking well, I'm 800' above ground lets check my vsi, ok I have enough time to retract the gear before impact lets pull it. You're thinking way to much into this!!! Simply unstall the wing with your yoke and throttle, this shouldn't take much longer than the 3 seconds it takes to relax the back pressure or put the nose a little forward and advance the throttle. Then when its no longer stalled pull the flaps to increase your climb performance, then check that you do in fact have a positive rate of climb and then pull the gear.

To leave the gear down for an obstacle 3/4 mile ahead seems pointless to me. The reduced climb rate with the gear doors is negligable and to keep the gear down for that obstacle only to clear it and hit the one behind seems pretty strange to me. Do you do that on a short field departure with an obstacle as well? Keep the gear down until you've cleared anything 3/4 mile off the end? What do you do in IMC? I've never flown anything with significantly more drag with the gear in transit, enough to worry about this. When the gear is half way up, you have less drag, doors or no doors, then when you have the gear completely down so you should be chopping that time down to 7 seconds. In other words when you have a positive rate of climb pop the gear to get a good climb rate and clear any obstacles ahead.

Really, like Hedley said, a stall is no big deal. I imagine he learned this when he started doing aerobatics. I learned this flying floats. When you're in a really demanding stick and rudder type of flying atmosphere you really learn the true effects of different ways that you can handle an aircraft. Up until that point it was a feared condition only recoverable by enormous amounts of lost altitude and great distress. Really, to stall an aircraft on approach, you should be able to recover without any significant altitude lost and still recapture the glide path and land. It shouldn't be much different than a downward air pocket. Hell if you're on an NDB approach and nowhere near the beacon crossing altitude, just stall it and you'll have a pretty approach

[justplanecrazy]

Goldeneagle as someone who's a CP and who normally I agree with, I was really surprised by your comments. I was hoping you'd reply so I could get a little more insight into your thought process. Maybe I misunderstood your post but I was more than a little surprised at your SOP's.

[Wacko]

I'm doing my multi atm as well, and on a seminole...turbo. Anyway... we bring the gear up first. I find that the gear creates a lot of drag.