Engine failure in multi

[SIGman]

Got my ME rating 3 years ago and haven't used it since
I'm sure I understand the 9 factors, etc, etc.... but I don't remember a good explanation of the following.
"
It doesn't make sense. I go through my whole Private, Insturment, Commercial, CFI, CFII (and twin, except engine failure) training under the impression that if the ball is centered, I'm coordinated. But wait a second.....Kill an engine in a conventional twin and the EVERY authority on multi-operations says that the rules change completely. All of a sudden, the way to stay coordinated is to fly what I had previously considered a sideslip (wings slightly banked, ball "split into the operating engine")

According to the FAA's Airplane Flying Handbook, some light twins will see as much as a 300 fpm increase in descent rate by holding wings level/ball centered. The way to hold coordinated flight is bank 2 degrees into the operating engine and center the yaw string with the rudder. (ususally end up with the ball slightly out of center)

Why the discrepency between the ball and yaw string? I can't remember it from my twin training. Theres a lot of experienced pilots on this board, how does it affect the aircraft you fly?

The only explaination I've recieved is that the rudder is creating a lateral force in the direction of the dead engine and it causes the sideslip. Last I heard, the rudder creates a torque around the Center of Mass that counters the torque effect of the operating engine. They cancel each other out. (the books say this creates a net sideslip towards the inop engine. . .HUH??)

whats the deal? Why can't I trust the ball to give me accurate coordination info with an engine failure?
"
If the above question doesn't make sense, I'll rephrase.

Cheers,
D

 

[Timebuilder]

You need to fly uncoordinated in a twin to eliminate the sideslip, which, as you recall, introduces a lot of drag. Banking into the good engine eliminates the slip, and therefore the drag, giving you the best single engine performance possible.

The info the coordination ball is giving you is accurate. You simply need to be doing something else!

 

[bobbysamd]

Eliminating sideslip during s/e ops

Great explanation, Timebuilder. Short and to the point.

You answered your own question, actually, when you said you were under the impression that the ball should always be centered, even in a twin, EXCEPT engine failure. That changes the rules.

Possibly another way to view it is that your thrust is unbalanced with an engine out, which is not the same as a single. Unless, of course, you're flying a push me-pull you Cessna Skymaster or a Do-336.

 

[surfnole]

Here's an example of what can happen if you don't handle engine failure properly

http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001207X03766&key=1

 

[DoinTime]

A good way to look at is that when you have your given angle of bank set in to eliminate slip you apply enough rudder to keep the yaw string centered. This will give you a uncentered inclinometer because you will be straight and level with a bank angle in. The ball will roll to the wing low side when in coordinated flight.

 

[DoinTime]

I forgot to mention.....
A good way to find out what your inclinometer should look like when flying engine out properly is to go out with both engines and put your airplane into a 5 degree bank (or whatever you like to use when single engine). Apply enough opposite rudder to get no heading change. The amount that the ball is off to the wing low side is the amount that should be seen when in coordinated single engine flight. Yes, you will be cooridinated in single engine flight.

 

[avbug]

Remember that 5 degrees is only a certification target. In reality, the required angle of bank will vary, and is often slightly less than 5 degrees. In some cases, at higher power settings in certain aircraft, it's slightly more.

 

[CooCooTim]

SIGman,
Why the discrepency between the ball and yaw string?

You've just deflected the rudder (ball centered) to stop the yaw and roll however, now you have a deflected rudder which is causing a sideslip. While sideslipping along, the relative wind is striking the vertical stab. and rudder combination at a decreased AOA. Also, the relative wind is flowing into the "side" of the vertical stab. thus blocking some of it from flowing over the rudder(It's striking the side of the whole plane too). Both of these things reduce rudder effectiveness. By banking toward the op. engine you are now producing a horizontal component of lift, "from the wing", the AOA of the rudder is increased and the relative wind is inline with the vertical tail which increases rudder effectiveness (decreasing VMC). As far as the ball displacement, it will be displaced more with the critical engine inop. Assuming it's a conventional twin. I hope this helps...As far as the control issue goes any way. This explanation is a lot easier with a diagram... Anyone have any thing else to add?

 

[Speedtree]

Let me address one small point possibly overlooked. You mentioned the torque of the dead engine counteracted by torque around the center of mass by the rudder. This is not an accurate way to view the situation. The failure of an engine produces yaw to one side, yaw produces roll to a certain degree which is where you find yourself when you lose the engine. You push rudder to correct for it. Rudder produces yaw, not torque, unless you want to view it as torque around the vertical axis which in my opinion is usually considered yaw.

It really is much better to look at the vector diagrams or some other visual diagram. Basically there is no opposite force to balance the horizontal force of the rudder input you are using to stop the assymetrical thrust of the one good engine. So, we use the small horizontal lift force created by the 3-5 degrees of bank to balance that and give us equal horizontal forces. The other explanations of why the ball ends up being slightly off are good.

One instructor gave an example of your twin sitting on a frozen pond. If only one engine is developing power and you apply opposite rudder to counteract (as if you are flying straight and level with ball centered) you would see your airplane sliding slightly sideways on the ice as it went forward.

I don't know if any of this helps but I hope you get it figured out. I know how it can be when you just can't get something to make sense in your own mind.

 

[Cornelius]

The methods discussed so far are correct for small twins and work great but larger aircraft are flown much differently than your typical seminole. Jets that use spoilers for roll control would be in serious trouble if you were trying to counter rolling tendencies with aileron/spoiler. It decreases the performance and increases sink rate big time because the spoiler on the operative engine side is sticking up in the air causing a lot of drag and loss of lift. The big boys use mostly rudder with a a continued takeoff above V1 to counter the engine failure. You'll also see that the ball is centered in flight, if it has a ball.

In the BE1900 when I was practicing V1 cuts, it took an incredible amount of aileron to counter the rolling tendency. Almost full scale aileron deflection at such low speeds was amazing and I think is common with V1 cuts in a lot of turboprops. Even turboprops a lot of times, you'll see the ball centered and not much bank in flight once you get your speed up. Remember the slower you go the greater the bank and rudder deflection you need to maintain the zero side slip condition.

 

[SIGman]

It helps clarify my confusion, I guess...

1. Am I understand that the ball (inclinometer) gives us "coordination" information but not necessarilly "slip" information? I've been taught and been teaching my students that they are the same thing. I thought coordinated = zero sideslip. No? not in a half busted twin? why?

2. Speedtree...you said what I said, I think. The operating engine (producing thrust) and inop engine (producing drag) produces a rotational force (torque) around the vertical axis (yaw). (this yaw also creates roll, I know)

We use rudder to create a rotational force (torque) around the vertical axis (yaw) to counter the engine failure. How much is required depends upon our KIAS and at least the "9 factors."

3. How do we fly the plane then? We have talked about (I assume) flying the aircraft at (or close to) Vmc. And have determined how to maintain the best CONTROL at Vmc. How about best perfomance at Vyse and Vxse? Same, Same??

Thanks for your responses and patience,
D

 

[DoinTime]

1. Am I understand that the ball (inclinometer) gives us "coordination" information but not necessarilly "slip" information? I've been taught and been teaching my students that they are the same thing. I thought coordinated = zero sideslip. No? not in a half busted twin? why?

Because your inclinometer is calibrated to give you coordination during wings level straight flight or coordinated banked turns. When you are in straight flight with wings banked (single engine operation) your inclinometer will always be wrong. I saw years ago an add-on adjustable inclinometer that you could calibrate yourself. You could set it for up to a 10 degree angle of bank. When you were single engine you would just adjust it to whatever angle of bank was appropriate and it would give accurate info. Pretty neat....haven't seen one since.

 

[Timebuilder]

You wrote:
1. Am I understand that the ball (inclinometer) gives us "coordination" information but not necessarilly "slip" information? I've been taught and been teaching my students that they are the same thing. I thought coordinated = zero sideslip. No? not in a half busted twin? why?

The meaning of the ball's position changes because it is responding to the inputs which became necessary during the single engine operation. The use of bank, which by the way reduces the rudder deflection from the initial amount, is the reason that the sideslip is being eliminated, and the ball wasn't designed to respond to this set of conditions. So, according to the turn coordinator, as designed, you are uncoordinated, but according to the airplane, which is the more meaningful indicator, you are coordinated because you have responded to the changing circumstances with changing control inputs, and eliminated the sideslip.


2. Speedtree...you said what I said, I think. The operating engine (producing thrust) and inop engine (producing drag) produces a rotational force (torque) around the vertical axis (yaw). (this yaw also creates roll, I know)
We use rudder to create a rotational force (torque) around the vertical axis (yaw) to counter the engine failure. How much is required depends upon our KIAS and at least the "9 factors."

The dead engine is responsible for yaw, and your first response is to apply rudder. The dead engine also creates roll as the lift is reduced on the side of the dead engine as induced airflow stops.
Flying this way, without banking into the good engine and keeping the ball centered, produces the performance robbing sideslip.


3. How do we fly the plane then? We have talked about (I assume) flying the aircraft at (or close to) Vmc. And have determined how to maintain the best CONTROL at Vmc. How about best perfomance at Vyse and Vxse? Same, Same??


We fly the airplane at blue line, or Vyse. Due to the factors, Vmc may indeed be lower than the red radial line, but I wouldn't go exploring that! Vmc is a useful guide to avoid a roll when low and close to terrain when density altitude is higher than your single engine climb ceiling. This means that you are in a controlled descent, and maneuvering to avoid objects as you prepare to set it down following your engine failure. The survivability of this scenario depends largely on avoiding the upside-down landing. Staying above Vmc is the way to avoid that rollover. How do you avoid this set of woeful circumstances? Good preflight planning is the answer. There are days when five or ten miles of clear, flat terrain would be necessary at the departure end of your runway to ensure a safe climb after an engine failure following takeoff on a warm day, or a controlled descent to an off-field landing on a hot day.

In other words, you can choose to not put yourself in a situation that doesn't allow you to safely climb after a failure and circle to land. Keep it on the ground whenever you can, following a failure. I was told this old chestnut once: it is better to go through the fence at 70 knots than to go through a barn at 150 knots.

I recommend these two FAA publications as required reading for my multi students:

FAA-P8740-19, Flying Light Twins Safely

FAA-P8740-25, Always Leave Yourself an Out


Think first, and then fly.
The fact that you are asking these questions is great. Good luck!

 

[avbug]

The inclinometer is in no way "calibrated." The closest that may be said an inclinometer comes to calibration is ensuring that the ball sits in the center when the wings are level on the ground, and that's a maintenance function of installing the instrument properly. Otherwise, there is no "calibration."

"Coordination" means using the controls of the airplane in such a manner that the airplane does what you want it to do. The common use term refers to keeping the skid ball in the center of the inclinometer, but this is erroneously applied. One coordinates the controls to make the airplane perform as desired. For passenger operations, the controls are generally coordinated such that the skid ball stays centered, but many operations require otherwise.

Note the purpose of the name "inclinometer." This is simply a primitive device which tells you which way something is leaning. In flight, we use it in conjuction with force vectors, intuitively; that is, the ball is acted upon by various forces in flight aside from gravity. When those forces are not acting on the ball in the inclinometer, it simply shows which way the airplane is tipped, or inclined, as gravity takes over.

A "dead" engine has little or no positive torque (often negative torque). Torque is manifest as a rolling tendency, proportional to the power produced by the engine, to a degree. It is not a yawing tendency.

"Coordination" has nothing to do with zero sideslip. It has everythign to do with making the airplane do what you want it to do. If the ball of the inclinometer is not centered, the airplane is not necessarily "uncoordinated." In the case of a multi-engine airplane with zero sideslip due to bank during an engine-out situation, the ball won't be centered, but this doesn't mean the airplane isn't "coordinated." It is coordinated. However, because the airplane is banked and isn't slipping or skidding, the ball will be course of gravity head downhill...typically a half to a full ball width in many light twins.

Imagine a single engine airplane in knife-edge flight. It is flying level, at a constant altitude, and is not turning, is not slipping, or skidding. Where is the skid ball? In the case of a left wing low knife-edge (90 degree bank), the ball is at the far left of the instrument. There it sits until acted upon by some other force than gravity. A similiar condition exists when operating engine out with a bank into the "good" engine in a multi engine airplane.

You teach side slips and forward slips. When coordinating the controls using opposite aileron and rudder (hard to accept the fact that cross controlling is coordinating, isn't it??) to slip, you'll notice that the skid ball is displaced opposite the rudder. That is, apply right rudder and left aileron to slip, you'll find that the ball moves off center to the left. Intuitively we know that to center the ball, we need only "step" on it, or use left rudder.

Imagine performing a side slip using a crosswind on final. A strong 90 degree crosswind is blowing, and you're banked into the wind to prevent drift. Where is your skid ball now? You're in a similiar situation to what you would be with an engine out on the downwind side of the airplane with respect to path over the ground (it's illustrative, not a perfect example). Aerodynamically there are differences, but by stopping your movement to the side, you will find that the ball of the inclinometer is off-center to the low side, much like a bank into a good engine in a multi engine airplane. Again, there are some differences aerodynamically, and this example, poor though it may be, is for illustrative purposes only.

Typically in large turbojet aircraft, the ball, when available (not found in all aircraft) is kept centered. However, not in all. I've flown several large types which required bank into the good engine(s) during engine-out work. The requirements vary with the airplane, not necessarily the size or powerplant. There are many large airplanes that are flown exactly as a seminole or other light twin.

 

[Timebuilder]

"Yea, wha he say."

 

[sstearns2]

.....

All engine out training in twins should be done with a piece of string taped to the center of the windshield.

The best way to look at this is to draw a basic top view of a conventional twin and draw the forces on it. Good engine thrust, airframe drag at the CG, and the force being generated by the vertical tail. Study this picture and you'll see that your flying sideways if you don't have any lift from the wing projected to oppose the force from the vertical tail.

The ball is centered if your wings level and your heading isn't changing. The ball just happens to have a close relationship to angle of sideslip in a centerline thrust airplane.

Scott

 

[DoinTime]

The inclinometer is in no way "calibrated." The closest that may be said an inclinometer comes to calibration is ensuring that the ball sits in the center when the wings are level on the ground, and that's a maintenance function of installing the instrument properly. Otherwise, there is no "calibration."

AVBUG...You contradict yourself here. Ensuring that an instrument that relies on being properly installed to give the desired results is in every way "calibrated." If it did not need to be "calibrated" you could bolt it on in any old way and get the desired effect.

Typically in large turbojet aircraft, the ball, when available (not found in all aircraft) is kept centered

What aircraft do not have a ball? or a digital representation of a ball (as in a glass cockpit)? A slip/skid indicator is required for IFR ops you know. 91.205.d.4

 

[avbug]

There is no calibration standard for an inclinometer. How many pounds of force, degrees of sideslip, g loadings, degrees of bank, or other measurement is given by ball displacement? Simply because an inclinometer ball rests at the low part of the glass at rest on the ground, doesn't make it calibrated. The inclinometer isn't calibrated. Ensuring that the item is properly installed isn't calibrating it. No mechanic calibrates an inclinometer. You won't find instructions or standards anywhere for calibrating an inclinometer. No manufacturer has standards or instructions for continued airworthiness for inclinometer calibration. Sticking it on the panel so it's straight is NOT calibration.

Not all large airplanes fly IFR, and not all large airplanes, or small ones for that matter, are bound by the minimum instrumentation spelled out in the FAR. Think about it.