Flying with 2 eng out

[74pilot]

We get the FIVE HUNDRED call on any approach, and it's part of the SOP.

I flew in a company once where all calls on approach were based on MINIMA and NOT on elevation. So there was

1000   (above the minimum, not RA nor elevation)
500
100     (above the minimum, not RA nor elevation)
Minimums  

But these were simple aircraft, so it was pilot calls. Why I mention it is that it actually makes so much sense. If all minimums were 200 feet AAD and below, it would not matter, but with some higher ILS minima, or some non-precision, you get the 500 call either between plus hundred and minimums, or even after! Some airports with serious obstacles have minimums well above 1000 feet AAD, and you will never get the 1000/500 feet warnings. If all these calls were referenced above MINIMUMS and NOT AAD/RA, then you would ALWAYS get these calls, at the right time.

Anyway, I am not running an airline, merely doing what I am told to do, hehe ... but in my spare time, I am free to air my thoughts about it.

Hardy, this thread was about 2-engines out, are you all clear about it now, or do you still have some questions about it? Sorry for deviating on the subject

[Qavion]

The 744 with enhanced GPWS can have what is known as a smart 500 callout.

(see page 2 of the following document)
http://egpws.com/engineering_support/documents/techpubs/1076-0976-34-16.pdf

During an EGPWS test, the "500" does not appear in sequence with the other callouts (this is a clue to its unique nature).

Hope this helps.
Rgds.
Q>

[Lasse]

Hey
(I know this was about 2 ENG OUT, but since it was brought up, and Im sure that Hardy or others will ask if more info on the 2 ENG OUT topic is needed...)

The EGPWS 500 "smart" call was introduced some years ago and you see that feature installed on the 787,777 and 737 as standart. Personally I dont like it, since the 500 call is used by alot of airlines as an indication of stabilized concept, but good thing that you always can pay a "little" amound of money and then... wula you can have it a "normal" standart call out for all approaches... Love those airline options...
The same with the rest of the call outs...Approaching Minimums, 100 above, Approaching Descion height well the list is long...

Cheers
Lasse

[74pilot]

OK, I guess our 500 callout is a dumb one then? It's not smart enough to know when it is ILS or not  :mrgreen:

[Hardy Heinlin]

I must dig out this old thread :-)

I learned from this thread that track and heading are equal when asymmetic thrust is compensated by the rudder. But I'm highly sceptical about the additional slip and bank theory.


When asymmetric thrust is compensated by the rudder so that heading and attitude are fixed, I can't believe that a non-zero bank angle will remain.

Why?

1. It would be impossble to use the ATT mode on the A/P. The ATT mode can keep only 0° or 5° and greater. When ATT is engaged while the current bank is, say, 3°, the A/P will roll to 0° and all the nice rudder trimming will be on the fritz.

2. Generally, for all AFDS roll modes, how could the AFDS know that a 3° bank angle in addition to the rudder compensation should be required? Normally, the F/D commands 0° bank when the desired heading is reached. If 0° bank would lead away from the command, the system would either oscillate or never reach the target. It would be unstable.


And when the wings are level and the heading is fixed, I can't believe that the ball will be out of center.

Why?

1. There's no centrifugal force on the yaw axis when the yaw axis stops turning

2. When the yaw axis stops turning, and track is equal heading, there's symmetric airflow on both wings, hence the wings are level, and therefore there's no gravity to one side to uncenter the ball.


I can't explain the sim photo in the first post of this thread. Why is the track zero while the ball is off-center? I'll check the video again. Perhaps the heading is not fixed in this moment and the aircraft turns around the yaw axis, that is, heading and track might equally rotate around the yaw axis while the yoke keeps wings level. That would generate a centrifugal force on the ball.


Cheers,

|-|ardy


Video checked. The heading rate is only 1°/sec.

[Hardy Heinlin]

Well, regarding the AFDS not knowing the asymmetic thrust condition: The AFDS roll modes could calibrate their 0° bank reference by watching that attitude at which the heading rate stops changing. When it stops changing at, say, 3° bank, then use 3° as the new wings level reference. I had something like that in my FD (outer loop) model before, but then I removed it because I think attitude and heading rates are under the control of the inner loop ("chose a velocity to move the aircraft to the commanded attitude") and not of the outer loop ("this attitude is your target").

...

As for the "unbelievable" ball offset: Of course, I actually believe it, I've seen it with my own eyes, all the big sims model it and everybody else has seen it. I just don't understand it. I can only explain it by assuming a tiny little bank angle or heading rate remaining when the aircraft is "stabilized". I'll model the half ball width offset, but it requires some tricks that I don't find elegant, it's like Copernicus' attempt to describe the Earth's elliptical orbit around the sun by adding multiple little circles.


|-|

[John H Watson]

As for the "unbelievable" ball offset:

Remember that the slip indicator has been adjusted by software over the years. What you saw in the sim 10 years ago may not be the same today.

Rgds
JHW

[Hardy Heinlin]

I think it's not software or aircraft related. I saw the "unbelievable" effect also on a 734 with a mechanical ball (sure, the mechanical ball is sim software driven, too, but ... ).


|-o|

[Will]

Doesn't the initial photo in this post show a track of 088 and a heading of 092?  That's quite a difference, isn't it, assuming no wind?

[Hardy Heinlin]

Yes, but that's probably influenced by the turn. On the second photo, with wings level, the track and the heading are equal. However, it's not 100% wings level. The ball might be left due to gravity. So, in the initial photo the turn might be coordinated, but it perhaps includes the same gravity component that is visible at 2° nearly-wings level.

If we need this nearly-wings level for the rest of the flight, the AFDS must recalibrate its roll modes.


|-|

[Mariano]

This will probably not help, but it might trigger further thoughts. From 767 training, I recall being taught (during engine out scenarios) to trim the rudder until the aileron scale on the yoke shows wings level and to disregard the slip/skid indicator.

Furthermore, use of aileron trim with autopilot(s) engaged is prohibited on the 767.

Extremely interesting and educational thread by the way. Only on Aerowinx's forum.

Mariano

[effte]

Hardy,
with asymmetrical thrust, you have a force couple in the horizontal plane between the center of thrust and the center of drag. To counter this force couple, you have the fin and rudder to create a lateral force.

In wings level flight, the only opposing force to help the lateral force from the fin/rudder create the force couple opposing the asymmetrical thrust force couple is the lateral aerodynamic force on the fuselage.  In other words, you have to yaw the aircraft enough into the good engines for the lateral force of the fuselage going sideways through the air to equal the lateral force of the fin/rudder. You will be in a slip aerodynamically, but wings level and without lateral acceleration so the ball will remain centred while the heading and track disagree. You'll be generating lots of drag through the side slip (beta angle), which is not ideal if you just lost a significant portion of your available thrust.

Your other option for generating an opposing force to the lateral force from the fin/rudder is to use the lift vector. The fin and rudder push the tail away from the running engines to counter the thrust asymmetry, so you have to tilt the lift vector into the running engines – bank into them.

Once in unaccelerated flight, you will have heading and track aligned, rudder into the good engine(s) and bank into the good engines. The ball will be where gravity dictates, i e earth-referenced straight down and towards the good engines in your airplane frame of reference. Quite hard to achieve perfectly, as your typical airliner does not have a tuft of yarn attached to the wind screen like real aircraft (i e gliders) do, nor a beta angle indicator. (I do recall the Concorde having a slip angle indicator.) I think the best you can do is to bank until heading and track line up.

Cheers,
   Fred

[Hardy Heinlin]

Thank you, Mariano and Fred, your comments help a lot.

Fred, what you're describing agrees exactly with my understanding of this whole principle. What confused me was probably the fact that those sim photos show such a small difference between track and heading. I expected a larger difference. I considered that small difference a non-difference and thus the whole principle became non-sense. -- I also expected that the ball wouldn't be so far off when the bank angle (gravity imbalance) is so small on the stabilized aircraft.

OK, if our conclusion is correct, the remaining problem is the AFDS. Let's take the HDG SEL mode as an example. Normally, when the selected heading is reached and the wings are level, the FD bar is centered. When I bank away from wings level, the FD bar goes off-center to the opposite side even if I'm not yet departing from the selected heading (inertia). The parameters for the FD bar are delta bank and delta heading. Not heading rate or bank rate which is handled by the inner loop. The FD bar is in the outer loop.

Before I continue, a quick excursion into the loops. This is how I understand the loop system (to keep it short, I just take HDG SEL as an example):


Outer loop -- the target bank angle:

This is a section in the flight control computer. In the outer loop the current bank angle, the current heading and the selected heading are continuously read (while also bank limits and other details are  considered, but let's ignore them for now, they affect not our problem). From these data, the outer loop computes a target bank angle that would lead the aircraft to the selected heading. This target bank angle is shown in form of a FD bar deflection. Rates are ignored. Heading rate and bank rate are ignored by the outer loop! When the FD bar wants me to roll from wings level to, say, 25° bank, the FD bar will approach the center as fast as my bank angle approaches the 25. When I bank very fast, the FD bar will not try to dampen me by moving the FD bar even faster to the center. The outer loop commands a bank angle only, not a bank rate. -- When the current heading gets closer to the target heading, the target bank angle gets smaller.


Inner loop -- the target bank rate:

This is another section in the flight control computer. This loop now does read the rate. It reads the current bank rate. And it reads the target bank angle (FD bar) commanded by the outer loop. From this, the inner loop choses a target bank rate and computes the respective yoke position that will roll the aircraft to the FD bar at the chosen rate. When the A/P is off, this job is done by the human brain. -- When the current bank angle gets closer to the target bank angle, the target bank rate gets smaller.


Servo loop -- the yoke movement:

This is yet another section in the flight control computer. It electrically adjusts the hydraulic valves so that the yoke is rotated to the position commanded by the inner loop. When the A/P is off, this job is done by human muscles. -- When the current yoke position gets closer to the target yoke position, the yoke motion rate gets smaller.


Back to the engine-out problem: Our aircraft is stabilized, current heading is fixed and agrees with the MCP selected heading, bank angle is 3°.

Now where's the FD bar?

Is it centered? Or is it deflected because it isn't happy with those 3° bank?

The AFDS has no data about asymmetric thrust. Both FMCs may be off. The only way for the AFDS to understand that the 3° bank is reasonable is this: Heading's OK and all rates are zero. Moving to wings level, i.e. setting a non-zero bank rate, would destroy the stable situation.

But the FD doesn't know the rates. Only the inner loop knows the rates.


Cheers,

|-|ardy



Edit: Outer/inner typos. I always confuse outer with inner, but only lingually, not technically :-)

[Hardy Heinlin]

Hello,

I started this thread 5 years ago. So, watch the year numbers when reading posts in this thread! :-)

Now ... one more thought regarding the 3rd photo below (I shot this in a LH sim):



I assume the track line is centered because the pilot maintains a bank of 3°. This slight bank also causes the slip indicator ball to deflect unusually far. The right wing goes down, and gravity pulls the ball to the right. Were the pilot to maintain wings level, the track line would be on the left side of the heading pointer (maybe -3° to -5°?).

In all photos I assume rudder is deflected correctly to stop the heading drift.


Regards,

|-|ardy


I will modify PSX later this year. Currently, PSX's ball is not centered when the heading drift is stopped by the rudder, prompting PSX pilots to overcompensate the imbalance.

[cagarini]

http://www.hoppie.nl/forum/var/2engOut.jpg

Because of the x-wind?  The photo was taken after an even more stable situation, and I managed to keep flying that way for quite a long time.

Thanks for the mention to the dates on this thread... as usual I started reading through it without checking dates...

[Hardy Heinlin]

What's your question, José?

[cagarini]

Not really a question, but rather showing that the ball was centered and heading / track  ( more or less) aligned, and I was not flying a curved path, so, it already works that way in PSX, I guess...

Or most  probably I didn't understand what you were saying here:

"I will modify PSX later this year. Currently, PSX's ball is not centered when the heading drift is stopped by the rudder, prompting PSX pilots to overcompensate the imbalance."

or it didn't apply to 2 eng out SITUs ? Or you meant when ** only ** the rudder is used ( no aileron input ) ?

[Xavier]

I'm sure you are already aware about this, however just in case :
Boeing's "747 Crew Training Manual" (aka FCT747(TM), 30/06/2010) specifically adresses this subject of adequate rudder trimming techniques and expected airplane behaviour, and actually answers some of the initial questions (from 2009). Pages 1.31 & 1.32

I preferred not to quote it there as it is under copyright, but the manual is available on the net...

[Hardy Heinlin]

José, I mean rudder only. Aileron neutral. Your stuff is centered because your aileron is deflected.


|-|

[cagarini]

Ok,

I guessed that after posting my previous message on this thread :-)

Anyway, if you didn't tell us you were going to update PSX to achieve it I would doubt it could  be possible, but I have to put my neurons to work....

Is there really an equilibrium state, with asymmetric thrust, where wings can be level and the "ball" centered, specially on an aircraft with such pronounced seepback, with just rudder input? No sideslip?