Singe engine stall in a twin near gross

A

Alexb2000

En-Route
Joined
Nov 28, 2010
Messages
3,530
Location
Dallas, TX
Display Name
Display name:
Alexb2000
I've been doing some homework on various twins and I came across a dangerous characteristic in at least some that I wanted to get some more information on:

The scenario is where you have an aircraft at or near gross (and aft CG) and lose one in a departure configuration. If in that circumstance the single engine stall speed is higher than VMC is there always the potential for a flat spin or is that model specific and what affects that characteristic?

Here is an example of a letter to the NTSB that describes the situation.

http://www.ntsb.gov/doclib/recletters/1981/A81_49_53.pdf

Ted, since I know you'll read this, how does the 310 stall on one at gross and near aft CG?
 
I'm guessing the flat spin isn't your biggest worry when a twin becomes a single at gross weight!:yikes: Vmc demonstrations are usually done with MEI in the right seat limiting rudder travel to avoid a Vmc rollover. :mad2:
I have 2000+ hours in multi engine airplanes and except for jets and turbo props you have a small window in which an engine failure is VERY, VERY BAD. The time from liftoff to when the gear is fully retracted is about 15 seconds in Charlene, that is a VERY critical time for everything to work. ;) For an engine failure at a low airspeed it is many times better to pull the other on back and land straight ahead.
Getting too slow in a twin is a recipe for disaster. :mad2:
 
Stick to single engine. Then when one packs it in as in your scenario, you know what is going to happen.
 
Note the date on the letter is 1981. Baron spin characteristics proved to be trickier than others of similar weight and size, and thought by many at the time to be the result of the clean wing that allowed the Bo-Baron planes to be such good performers.

VMCA is calculated at max weight, aft CG, full power on one engine with the other feathered and secured. Actual loss of control can occur at other speeds if any of the conditions are changed. The after-takeoff accidents are usually the result of loss of control due to VMC, with insufficient time for a flat spin to develop.

You would probably enjoy seeing such action first-hand in a sim.
 
N747JB said:
I'm guessing the flat spin isn't your biggest worry when a twin becomes a single at gross weight!:yikes: Vmc demonstrations are usually done with MEI in the right seat limiting rudder travel to avoid a Vmc rollover. :mad2:
I have 2000+ hours in multi engine airplanes and except for jets and turbo props you have a small window in which an engine failure is VERY, VERY BAD. The time from liftoff to when the gear is fully retracted is about 15 seconds in Charlene, that is a VERY critical time for everything to work. ;) For an engine failure at a low airspeed it is many times better to pull the other on back and land straight ahead.
Getting too slow in a twin is a recipe for disaster. :mad2:
Click to expand...


At gross in Charlene is the single engine stall speed above Vmc?

BTW- she's looking good.
 
N747JB said:
Vmc demonstrations are usually done with MEI in the right seat limiting rudder travel to avoid a Vmc rollover.
Click to expand...
That is completely incorrect.

VMC demos are often done with the MEI limiting rudder travel in order to simulate the loss of control simply due to the fact that these days because we do our engine failure training at altitude, the actual VMC at altitude is typically well below stall speed in the majority of light trainer-type twins.

If you were to try a real VMC demo without limiting rudder travel in a Travel Air or Duchess for example at 5000', you would stall the airplane well before you ever reached VMC.
 
This is an interesting question. It doesn't surprise me that the Baron series has this characteristic given my flight time in an A36 with a radar pod on the wing. The plane always dropped that wing abruptly in a full stall.

Multi-engine training has changed a good bit because of the realization that much of what was standard training practice caused more crashes than the event itself in the real world. In the old days, you had to demonstrate ability to recover from single engine stalls. I've stalled the Aztec OEI (intentionally) and it handles it nicely, just like everything else stall related. Remember that you'll typically have an OEI stall when at a part power setting, since that will lower Vmc. Naturally aspirated twins will be more susceptible since their power decreases with altitude as soon as you're off the ground.

The 310 I fly has VGs on it, which in my experience with the plane effectively eliminate Vmc, even with the bigger engines and 3-bladed props. Intentional attempts to make a Vmc roll occur have produced no results. Stalls are quite docile. But I also don't try them when I'm at gross weight with an aft CG near limits because I don't want to tempt fate. This gets into my next point.

The real answer in practice is just don't get there. Blue line is the minimum airspeed you should ever see OEI. Obviously people screw up and don't do that, but it's less common than the NTSB reports imply. Most engine failures in twins result in a safe landing with no resulting report. If you don't allow yourself to get there, problem solved. This means that on takeoff like John said, there may be a time during which an engine failure means you pull both back and land, even if it's off-field. Depends on your plane, but a proper twin pilot needs to have decided at any point what the immediate response to an engine failure is.

Blue line should be treated as a speed to never get below when OEI, and should be treated with caution when you have two turning. I have to get below blue line on final to make it into my home field. Taking off I'm above blue line by the time the wheels are up (power is wonderful). That's a risk I take and I consider it minimal due to the short exposure time and the fact that I'm on a very low power on approach.

So how do you handle this practically to minimize risk? Use long runways, and always stay above blue line. If you just don't get there, you won't crash.
 
Alex, the date on the NTSB report is old. You really cannot say for certain that any aircraft max gross stall speed is higher that VMC because VMC is always changing with conditions. Tower is right that at altitude where we actually practice VMC demos that is were the issue comes into hand of stalling before VMC is reached. At a high power setting and asymmetrical thrust its gonna be an issue. The whole point in that letter is to describe the "unusually violent" tendencies of the travel air to spin in these conditions. In a takeoff situation, you are sometimes fighting a losing battle trying to make a crippled twin fly. And like Ted said, sometimes you need to pull them both back and go strait ahead. But there has been hundreds of conversations about that.
 
For those that have been flying twins and training prior to the letter (1981), I am curious if that was the impetus for changing the ME PTS to stop the VMC demo at the stall warning or buffet if they occur prior to actual loss of directional control?
 
wabower said:
Note the date on the letter is 1981. Baron spin characteristics proved to be trickier than others of similar weight and size, and thought by many at the time to be the result of the clean wing that allowed the Bo-Baron planes to be such good performers.

VMCA is calculated at max weight, aft CG, full power on one engine with the other feathered and secured. Actual loss of control can occur at other speeds if any of the conditions are changed. The after-takeoff accidents are usually the result of loss of control due to VMC, with insufficient time for a flat spin to develop.

You would probably enjoy seeing such action first-hand in a sim.
Click to expand...

Wayne-

I would really like to see that. It's interesting to me and most MEI's want to keep the OEI work as benign as possible so you never really discuss those scenarios.

I saw the date, I just didn't know if something had changed and/or only effected those models.
 
It was simply a "whichever comes first" rationale based on the knowledge that either event might occur before the other. One of the reasons that the Duchess became so popular as a S/E trainer was that it had been successfully spin-tested OEI.

Fearless Tower said:
For those that have been flying twins and training prior to the letter (1981), I am curious if that was the impetus for changing the ME PTS to stop the VMC demo at the stall warning or buffet if they occur prior to actual loss of directional control?
Click to expand...
 
Until the rash of accidents nobody really knew that the family of planes were so nasty. At one time the training and testing included simulated engine failures at low altitudes, but that too was discontinued due to the dwindling pilot population resulting from such activities.
Alexb2000 said:
Wayne-

I would really like to see that. It's interesting to me and most MEI's want to keep the OEI work as benign as possible so you never really discuss those scenarios.

I saw the date, I just didn't know if something had changed and/or only effected those models.
Click to expand...
 
Alexb2000 said:
Wayne-

I would really like to see that. It's interesting to me and most MEI's want to keep the OEI work as benign as possible so you never really discuss those scenarios.

I saw the date, I just didn't know if something had changed and/or only effected those models.
Click to expand...

I think the issue with in the aircraft training is that it's become more focused on exposing students to what can happen without putting them in situations where they have a high probability of actually killing everyone on board.

I'd like to do the sim work as well. Sounds like a cause for all of us to get together in Dallas.
 
Ted DuPuis said:
I think the issue with in the aircraft training is that it's become more focused on exposing students to what can happen without putting them in situations where they have a high probability of actually killing everyone on board.

I'd like to do the sim work as well. Sounds like a cause for all of us to get together in Dallas.
Click to expand...
How well do the sims model VMC and VMC rollover?

I've seen videos of VMC rollover and would prefer to never experience such an event in an actual aeroplane.
 
flyingmoose said:
Tower is right that at altitude where we actually practice VMC demos that is were the issue comes into hand of stalling before VMC is reached. At a high power setting and asymmetrical thrust its gonna be an issue.
Click to expand...

So if one were to operate a twin from mountain airports, same or higher than the altitude Tower suggested is the stall now more relevant than Vmc for all twins or just some like the Beech's in the article?
 
Fearless Tower said:
How well do the sims model VMC and VMC rollover?

I've seen videos of VMC rollover and would prefer to never experience such an event in an actual aeroplane.
Click to expand...

That sounds like a question for Wayne.
 
Alexb2000 said:
So if one were to operate a twin from mountain airports, same or higher than the altitude Tower suggested is the stall now more relevant than Vmc for all twins or just some like the Beech's in the article?
Click to expand...

Probably more relevant for non-turbo planes, especially those Beech models.
 
Ted DuPuis said:
I'd like to do the sim work as well. Sounds like a cause for all of us to get together in Dallas.
Click to expand...

If it's not a private party, I'd like to join in! As a relatively recent MEI, I'm very interested.
 
Alexb2000 said:
No change in your bird's SOP for OEI in the mountains?
Click to expand...

Cessna didn't publish any special considerations for mountains, and I don't see why I'd do any differently from my flatland techniques, since I still want to maintain blue line or better at all times and never below it. The obvious difference for me, lacking turbos, is that my takeoff roll and acceleration will be worse, so if I have a danger zone it will be larger than here in the oxygen-rich environment of Ohio.
 
Alexb2000 said:
No change in your bird's SOP for OEI in the mountains?
Click to expand...
I don't think they publish that kind of procedure, because VMC is kind of a feel thing. If you are in that situation, you have to be able to fly that particular airplane in those situations. Chances are good that if you are seeing signs of VMC, you shouldn't be where you are and should think about doing something else. Now if you are starting to see signs of stall before you VMC, then you also need to change what you are doing. Doesn't matter what happens first, you may not be able to fly it in that situations.
 
flyingmoose said:
I don't think they publish that kind of procedure, because VMC is kind of a feel thing. If you are in that situation, you have to be able to fly that particular airplane in those situations. Chances are good that if you are seeing signs of VMC, you shouldn't be where you are and should think about doing something else. Now if you are starting to see signs of stall before you VMC, then you also need to change what you are doing. Doesn't matter what happens first, you may not be able to fly it in that situations.
Click to expand...

Exactly. And if you follow procedure correctly, you shouldn't get there.
 
flyingmoose said:
I don't think they publish that kind of procedure, because VMC is kind of a feel thing. If you are in that situation, you have to be able to fly that particular airplane in those situations. Chances are good that if you are seeing signs of VMC, you shouldn't be where you are and should think about doing something else. Now if you are starting to see signs of stall before you VMC, then you also need to change what you are doing. Doesn't matter what happens first, you may not be able to fly it in that situations.
Click to expand...

Sounds like something that should be practiced in some manner? How else do you keep the feel sharp?

Here is a link to a tragic Baron crash in OK this January. From the picture it looks like almost no forward motion and flat attitude, which to me suggests a flat spin?

http://newsok.com/men-who-died-in-mangum-plane-crash-loved-flying/article/3747898
 
Alexb2000 said:
Sounds like something that should be practiced in some manner? How else do you keep the feel sharp?
Click to expand...
There are some things which should be avoided rather than practiced unless you do it in a sim. You don't need to practice bleeding. I think that some people get the mistaken impression after flying single-engine trainers that everything you could do in an airplane is going to be easily recoverable. When you get into higher performance airplanes, such as twins, this may not be the case.
 
All of the roll-over characteristics displayed by the sims are sufficient to dissuade a pilot from going home and trying them himself in the airplane. Whether they are absolutely accurate isn't something I want to know badly enough to find out through actual experience. I've read enough NTSB reports to know all I need to know about that subject.

The sim designers say the flight characteristics within the known envelope are very accurate, and those at the edge are based on engineering data from mfr test flights and other known sources. I'm not qualified to opine on the accuracy of these statements.

Interestingly, the most violent of all such maneuvers in the sims I've flown has been the yaw-damp hardover in the Citation 650. I've yet to see any pilot successfully recover, since the yaw-damp engage command is usually the last of three issued ("gear up, flaps up, yaw damp") during climb, and shortly after takeoff is completed. Pilots who have experienced the smoking-hole outcome, however, usually have a firm grip on the controls when they call for the Y/D during future flights.

Ted DuPuis said:
That sounds like a question for Wayne.
Click to expand...
 
Everskyward said:
There are some things which should be avoided rather than practiced unless you do it in a sim. You don't need to practice bleeding. I think that some people get the mistaken impression after flying single-engine trainers that everything you could do in an airplane is going to be easily recoverable. When you get into higher performance airplanes, such as twins, this may not be the case.
Click to expand...

This brings up a point I was about to make, which is that in aviation we are full of limitations and realms of flight we avoid because of safety consequences. We don't exceed Vne or Mmo. Turbine Commanders aren't supposed to exceed 180 KIAS in moderate turbulence (this is separate from Va), you get the idea.

One former U-2 pilot I met who owns a Lancair IV-P said he stalled it once. It did a split S on him. He recovered, added an AoA indicator and said "I just don't do that anymore."
 
I think being as far under MGTOW would be of great help too. Look at the book number for that in your airplane then subtract say #200 and make that your own new number. Yeah, it cuts into utility but helps out with worst case. YMMV.
 
Fearless Tower said:
That is completely incorrect.

VMC demos are often done with the MEI limiting rudder travel in order to simulate the loss of control simply due to the fact that these days because we do our engine failure training at altitude, the actual VMC at altitude is typically well below stall speed in the majority of light trainer-type twins.

If you were to try a real VMC demo without limiting rudder travel in a Travel Air or Duchess for example at 5000', you would stall the airplane well before you ever reached VMC.
Click to expand...

Actually, in a Travel Air you don't need to be much above 3000'.
 
The amount of built-in help available to the pilot can be a significant determinant insofar as the successful outcome of OEI maneuvers is concerned.

A high percentage (but not all) of the King Air fleet has both rudder boost (provided by bleed air) and auto-feather (activated by torque sensors) that assist the pilot during engine failure. During "batting practice" day in the sim, pilots are allowed to practice all types of VMC maneuvers as well as single-engine missed approaches from both M/E and S/E approaches. The success rate is near 100% with both of the assist systems operational, but drops significantly when one or more are disabled by the IP.

The lesson for pilots who operate twins without either assist system is that during an engine failure they will 1.) be be very busy and 2.) need to make each movement correctly the first time in order to survive.

Ted DuPuis said:
This brings up a point I was about to make, which is that in aviation we are full of limitations and realms of flight we avoid because of safety consequences. We don't exceed Vne or Mmo. Turbine Commanders aren't supposed to exceed 180 KIAS in moderate turbulence (this is separate from Va), you get the idea.

One former U-2 pilot I met who owns a Lancair IV-P said he stalled it once. It did a split S on him. He recovered, added an AoA indicator and said "I just don't do that anymore."
Click to expand...
 
The long and short of it, don't get that slow, and if you do, keep your hand on the good one and be ready to pull it back to maintain control. You can't always climb and when you can't, you just take what you get, keep it under control and pick the best place to stick it. The scenario in the OP though brings up a good point about 'Max Gross Weight', try not to operate there in a light twin. Just 10% under buys you a pretty nice margin. Normally I'm flying with half my useful load unused at take off, that buys me rather impressive OEI performance for a recip twin.
 
wabower said:
The amount of built-in help available to the pilot can be a significant determinant insofar as the successful outcome of OEI maneuvers is concerned.

A high percentage (but not all) of the King Air fleet has both rudder boost (provided by bleed air) and auto-feather (activated by torque sensors) that assist the pilot during engine failure. During "batting practice" day in the sim, pilots are allowed to practice all types of VMC maneuvers as well as single-engine missed approaches from both M/E and S/E approaches. The success rate is near 100% with both of the assist systems operational, but drops significantly when one or more are disabled by the IP.

The lesson for pilots who operate twins without either assist system is that during an engine failure they will 1.) be be very busy and 2.) need to make each movement correctly the first time in order to survive.
Click to expand...

I'd agree with this, and I've thought about how one could implement an auto-feather setup in a piston twin. That by itself would make a huge improvement. I hadn't thought about rudder boost, but that would be another good one to add.

Of course, I don't see either coming around anytime soon to this realm.
 
Ted DuPuis said:
I'd agree with this, and I've thought about how one could implement an auto-feather setup in a piston twin. That by itself would make a huge improvement. I hadn't thought about rudder boost, but that would be another good one to add.

Of course, I don't see either coming around anytime soon to this realm.
Click to expand...

Auto Feather would not be difficult to build, torque sensor on the engine mount running a dump valve on the prop oil pressure. Getting it certified however....:rolleyes2:
 
wabower said:
VMCA is calculated at max weight, aft CG, full power on one engine with the other feathered and secured.
Click to expand...
23.149(b) says otherwise, unless you take off with the prop control in the feathered position (unlikely), or the prop autofeathers on loss of torque (as do some turboprops). And while aft cg is usually the worst case, the reg just says "most unfavorable position," which I suppose might be something else if the aircraft is very unusually designed.

As for ME training, the PTS says the Vmc demo is done at the bank angle for best performance and control, not the 5 degrees specified in 23.149 to determine "book" Vmc during certification. In any light twin out there, that is going to be within a fraction of a degree of 2 degrees of bank into the good engine, not 5, which trades off a lot of performance without gaining significant control advantage (because Vyse is so far above Vmc). By using 2 degrees of bank, Vmc is raised significantly in relation to stall speed, allowing the desired effect (yaw which can't be controlled by available rudder requiring the trainee to reduce power on the good engine to maintain directional control) without the instructor having to mess with the rudder pedals. If necessary, you can even go to zero bank to achieve that goal.

For more on this, see the late Prof. Mel Byington's excellent paper on point at http://robyadougacfi.com/files/Engine-Out_Booby_Traps.pdf. And if anyone knows where one can find his superb video "Optimized Engine-Out Procedures for Multi-Engine Aircraft" on line, please let us know. Otherwise, you'll have to order the original VHS tape for $28 from:
Embry-Riddle Aeronautical University
University Distribution Center
Daytona Beach, FL 32014
(904) 239-6484
​
 
Last edited:
Henning said:
Auto Feather would not be difficult to build, torque sensor on the engine mount running a dump valve on the prop oil pressure. Getting it certified however....:rolleyes2:
Click to expand...

The catch would be calibrating it so it didn't feather when windmilling. Easier in experimental.
 
Fearless Tower said:
For those that have been flying twins and training prior to the letter (1981), I am curious if that was the impetus for changing the ME PTS to stop the VMC demo at the stall warning or buffet if they occur prior to actual loss of directional control?
Click to expand...

My DPE for the ME checkride was talkig about this. He owns a few Apaches of his own and one day he and an aerobatic pilot got permission to test a single-engine stall in one. Apparently it did spin, and the experienced aerobatic pilot lost over 5000' altitude before finally recovering.
 
FWIW, when Grumman developed the GA-7 Cougar, they initially intended to certify it for intentional spins, and got part way through the spin certification test program (without any hitches) before someone decided that was a waste of money. I heard Roy LoPresti, who was the chief designer on the project, talk about that some years back before he died.
 
Alexb2000 said:
DA42 auto feathers and it works really well. The technology is out there.
Click to expand...

On the diesels only or also the Lycoming variant? With the diesels it'd be easy to program in the FADEC, which I think is the way to do it.
 
Ted DuPuis said:
On the diesels only or also the Lycoming variant? With the diesels it'd be easy to program in the FADEC, which I think is the way to do it.
Click to expand...

I believe you're right, how long until FADEC on gas motors, sheesh.

Everskyward said:
There are some things which should be avoided rather than practiced unless you do it in a sim. You don't need to practice bleeding. I think that some people get the mistaken impression after flying single-engine trainers that everything you could do in an airplane is going to be easily recoverable. When you get into higher performance airplanes, such as twins, this may not be the case.
Click to expand...

I don't believe I'm talking about anything out of the ordinary flight regime here. Piston twin leaves higher altitude/DA airport, loses one, pitches for Vmc, stalls on one trying to get engine secured, adios. If I have that right, it seems like a training failure that it isn't even discussed as part of OEI work in many cases. Also, as nice as SIM training would be most non-pressurized light twin pilots don't do that or it isn't available for their type. :dunno:
 
You must log in or register to reply here.
Back
Top