Automatics versus flying skills - Are some pilots scared to fly by hand?

[Mshamba]

I am an "ab initio" pilot and working as an FO now since half a year. I do so well remember my very first "real" flight after the typerating, my first supervision flight.

I made the takeoff, as we had the acft clean and our 250 knots, 3000 ft height or so, i turned on the AP - as used from the SIM. My supervision captain switched everything off immediately. AP off, FD off - and even my beloved magenta, leaving my flying manually all the way up to cruising level. 10 minutes after cruising level he allowed me to switch on the automatics.

Descend: 10 miles before top of descend the same story. All switches off, and me, the poor desperated FO who's mind was still on the gate had to manage to fly the whole approach manually - raw data. Well - i did manage and despite i cannot remember every detail i'm quite sure i managed it well.

After this experience i was cured. I learned most probably the biggest lesson of my flying-life: you CAN fly manually, even if you are a bloody beginner - the whole flight. You don't need any automatics! Today i laugh about that - but at that very day it was new for me.

I am quite sure most of the pilots fresh out of the factory number next do not expire such a supervisor, nor do they fly that much manually. So where should they get this so desperately needed self confidence from? If you never experience a manual flight from the beginning to the very end, if you always use FD/AP/AT/magenta line - no wonder you are "trained" to be frightened by any manual and/or raw data flying. The reaction of the quoted frightened FO's are understandable. And my kind advice: drop them into the cold water and let THEM fly manually, that they get their self confidence what they obviously do not have.

I am thankful 'til today to my supervisor for that lesson (thanks Gerry ) - and love to fly manually as much as i am allowed by my "magenta-loving" captains...

PS: I experience quite a lot captains who fly the automatics from 500 ft to 500 ft... even a visual approach must be put into the FMC to be flown... asked about the answers they're always the same: for economical reasons, to see deviations from the descend path, better management in general or the beloved sentence "i flow more manually in my life than all other pilots put together, so now i enjoy the automatics". Cheers, dude.

[Atlantean1963]

Please excuse an SLF jumpingin here, but from an SLF perspective, the response to the original question

"Are some pilots scared to fly by hand?"

can only be to point out that if the pilot is scared to fly by hand, then there are several hundred people sat behind him who are bloody terrified...

I suspect that as SLF we're not really bothered if you do or you don't fly by hand - but we really, really, really, want to feel confident that you can if you have to...

Best Regards,

Atlantean.

[Tee Emm]

and flying through the flight director becomes nothing more than simply utilizing all the resources one has before him rather than blindly following only one.

A re-occuring event I often see during simulator training (B737) is the difficulty of some pilots when conducting a single engine go-around and trying hard to follow the flight director commands. At the moment of go-around, these chaps press the TOGA button which normally causes the FD needles to guide the pilot to the correct roll/pitch angles. In the 737 on a single engine approach the autothrottle is switched off (Boeing recommendation).

But what sometimes happens is the pilot presses TOGA but momentarily forgets to push open the thrust lever because in his two engine go-arounds the TOGA button also operates the autothrottles (if armed of course) and the pilot is used to instant thrust application from one push of a button. By now the FD needles are all over the place and the grim chase to centre them is on. Even if the pilot does push the live engine thrust lever to the go-around N1, the resulting yaw and roll, if not immediately countered correctly, leads to more FD needle gyrations and I have observed significant angles of bank accompanied by significant changes of heading as the pilot concentrates his efforts on trying to satisfy flight director signals.

The point made of "seeing beyond" the FD needles or flying "through" the FD is valid. In fact, in preparation for the early simulator lessons on single engine go-arounds, one teaching method that works remarkably well is to first practice the go-around procedure raw data - that is FD off. After the first couple of attempts using basic ADI scanning, where the actual planned body angle can be readily seen on the ADI, unhindered by two needles jerking in seemingly random directions while blocking the view of the "little aeroplane", we found that pilots could do a perfectly satisfactory job of a go-around.

One good piece of advice offered to pilots who find the FD needles a hindrance in a single engine GA, is to scan the standby ADI and use that instrument to keep wings level and body angle pinned until the FD needles settle down and give you more precise commands.

The point I am making is that in the iniitial flurry of activity associated with a single engine FD go-around, we see more excess bank angles, heading changes and varying of airspeed, when eyes are glued to following FD commands, than in the theoretically more demanding raw data go-around. Of course a pilot should be equally competent at both types of single engine go-around. But, that seldom happens - especially where recurrent simulator training is months apart and most of that is on automatics anyway.

[Denti]

A re-occuring event I often see during simulator training (B737) is the difficulty of some pilots when conducting a single engine go-around and trying hard to follow the flight director commands. At the moment of go-around, these chaps press the TOGA button which normally causes the FD needles to guide the pilot to the correct roll/pitch angles. In the 737 on a single engine approach the autothrottle is switched off (Boeing recommendation).

Boeing made some changes in its later models to cater for the automation folks. Since CAT IIIa single engine approaches, which includes automatic go-around capability, is now available it is of course easier to fly the one engine out go-around (because the automatics fly it), however it is allways very interesting to see the gyrations once a different vertical mode is selected and the automatic rudder kicks out and it is the pilots responsibility to care about yaw and associated stuff. I certainly don't want to sit in the last row of any airplane with an untrained crew where that happens

[Pugilistic Animus]

I liked to watch when folks used the FD on the 757/767 as a "TO director" in the sim a nasty surprise may await---you'll learn to fly through it

the Cirrus aircraft's G1000 is maybe 100X more capable than that of the Boeing amazing but
I still have not checked out you need about 25 hrs to fly and that's VFR-only ---IFR--I don't know---give me a Mooney or a Bonanza sooo much easier---and yes one can make real compass turn if you know your latitude

[oxenos]

The latest report on the Quantas Airbus emergency landing ( R & N ) states that the pilots disengaged the auto pilot and hand-flew the aircraft for all but a few seconds following the initial problem. I am sure no-one would question their wisdom in doing so. How well would the "auto-pilot in at all times" brigade have coped?

[Carnage Matey!]

Probably equally well given that it's a fly-by-wire Airbus!

[GearDown&Locked]

How well would the "auto-pilot in at all times" brigade have coped?

See the example of the BA038... as far as we know, A/P only disconnected by itself when out of operational range. If we remove the fact that the A/C was too close to the ground to allow other options and if it ocurred at 2000ft AGL would hand flying the A/C made any difference to the outcome?

And to the non-magenta pilots: Would you have felt safer/more confident shutting off all automatics and hand-flown the plane to the ground on a similar situation?

Excellent discussion between all of PPrune's heavy weights BTW.

GD&L

[SNS3Guppy]

One of the reasons I `ve remained a military pilot. I want to be able to hand fly when the circumstances warrant it, and not be a glorified computer programmer/autopilot monkey.

Your assertion,then, is that nonmilitary pilots can't fly, or are "glorified computer programmer/autopilot monkeys?

[Stanley Eevil]

One of the reasons I `ve remained a military pilot. I want to be able to hand fly when the circumstances warrant it, and not be a glorified computer programmer/autopilot monkey.

Your assertion,then, is that nonmilitary pilots can't fly, or are "glorified computer programmer/autopilot monkeys?

You may think that; I couldn`t possibly comment.

[BelArgUSA]

Stanley -
xxx
What do you expect in this day and age of MCL licences...?
Takes a lot of training, practice and airmanship to move an "ON & OFF" switch...
xxx
Somewhat similar to that American lady last month in Madrid.
I was in a car hire office... She asks me "do you speak Spanish...?"
I say yes...
"Could you ask them to give me an automatic transmission car instead...?"
She could not drive a regular transmission...
xxx

Happy contrails

[Loose rivets]

Odd that. A couple of life-long lady friends at home, have never held a license that entitles them to drive a manual transmission.


I suppose we were lucky that there were no simulators about when the first 1-11s were delivered. Upper air work included aileron through to the spoiler, and stalls through to the push. Quite innocuous really, but I would hate for it to have happened on a 'dark and stormy' without having practiced it first. Well, I'd hate for it to happen then anyway, cos of the tea and bikkies conversation that it would have prompted, but you know what I mean.

The memories come flooding back. 1500' QFE over the threshold at SEN in a Viscount, then being told 'Now land it.' And at Teeside training on the new 1-11 doing the bad weather circuits. Just fabulous fun. One liberal type let me charge round at 500' on full noise with 60 deg bank. I can still see him, way up above me, laughing his head off. When did aviation stop being fun?

My attempts at doing BW circuits in the sim, were pathetic. Just guessing angles in the haze really. No comparison.


I don't know what the solution is. Even if all the airlines pooled their resources, and tried to follow Davis' logic, it would still be too expensive to run an aircraft large enough to be of use. That is enough aircraft for crews to all get a go at real handling. Not being on type wouldn't matter for this detail, first and foremost it would be the g-forces during these maneuvers that would be the main difference. When things go wrong, simulated gravity sucking at you from the wrong directions is very disconcerting. A sim that could replicate that would be a trick, wouldn't it?

[Denti]

Well, there is at least one simulator that can simulate some g-force for longer than the usual level D simulator can. But i guess access to it is not easy.

NASA Ames Aviation Systems Division: VMS

[SNS3Guppy]

One of the reasons I `ve remained a military pilot. I want to be able to hand fly when the circumstances warrant it, and not be a glorified computer programmer/autopilot monkey.

Your assertion,then, is that nonmilitary pilots can't fly, or are "glorified computer programmer/autopilot monkeys?

You may think that; I couldn`t possibly comment.

You already did comment, you see. You said it. Clarification sought, none given, your statement thus stands.

Simply because one is not a military pilot does not make one "glorified computer programmer/autopilot monkey," and believe it or not, airplanes get hand flown outside the military, too.

[Tee Emm]

There is a simple and effective solution to all this. The cyclic, recurrent and type rating simulator sessions should include a significant proportion of hand flying without the aid of automatics and that includes the flight director. Touch and go circuits and landings are very good for sharpening instrument scan on basics. Of course there are very obvious limitations to "practicing" on revenue flights, and commonsense should prevail. Switching off a flight director in flight is not exactly dare-devil stuff - although there are many who shudder at the thought.

The majority of simulator practice is on automatics when already 95 percent of actual flight is also on automatics. If the ideal is to be one hundred percent competent at both skills (automatic monitoring or basic handling), then equal time must be allotted in the simulator. Inordinate time spent taxiing a simulator to a holding point a mile away is time wasting. A landing configuration stall at 500 feet on final is not a time waster - nor is a stall recovery at 37,000 ft. These have happened yet the only time stall recovery is practiced is in initial type ratings. Even then the landing configuration stall is usually limited to 5000 ft agl for goodness sake.

Some regulatory authorities permit unrestricted use of automatics for an instrument rating test. A hand flown steep turn is about the only "test" of manipulative skills. Yet the tolerances allowed in the instrument rating test remains whether the automatic pilot can fly "within tolerances" or the pilot.

Simulators are invaluable devices for training pilots to fly aeroplanes. But if it is agreed that a high level of competency is required of flight crews both in automatics and basic manipulative skills, then operators need to stop just paying lip service to the need for basic flying skills and schedule adequate simulator time to cater for this.

[misd-agin]

Geardown&locked - See the example of the BA038... as far as we know, A/P only disconnected by itself when out of operational range. If we remove the fact that the A/C was too close to the ground to allow other options and if it ocurred at 2000ft AGL would hand flying the A/C made any difference to the outcome?

And to the non-magenta pilots: Would you have felt safer/more confident shutting off all automatics and hand-flown the plane to the ground on a similar situation?
********************************************************

The question could also be asked if hand flying would have made the incident worse? I say no, hand flying would not make it worse. The worst they could have done hand flying was to slow to the stickshaker and then have the automation kick in while they fell to the ground.

[BelArgUSA]

Simulator - Recurrent - Captain 747

I had, one day, having only 2 hours of simulator available...
And 3 captains to give a recurrent 6 month check.
Of course I knew each of them as competent pilots.
So I did each of them, in what I called a "quickie"...!
xxx
Located the simulator "on the runway" engines running, short training check list completed.
Conditions 800 meters visibility, ceiling 200 feet, crosswind 10 knots.
Weight selected at maximum landing weight...
1. First takeoff, engine failure at some 5 knots below V1... Rejected T/O...
Reposition at takeoff on runway, all engines running again.
2. Second takeoff, outboard engine failure at V1... Continue T/O.
Climb to 800 AGL, level off, flight director/autopilot inoperative. Complete check lists.
Request return to airport. Reposition downwind about to turn base.
3. Engine-out ILS down to minimums, visibility/ceiling zero-zero. Missed approach.
4. Engine fire inboard same side as failed engine, during missed approach.
Climb to 800 AGL. Complete check lists, request return, reposition to downwind.
Ceiling now up to 600 feet, visibility is good.
5. Two-engine NDB approach, using the OM...
6. Landing to full stop. End of check ride.
xxx
The above takes 30-35 minutes to complete.
What else could I dream to ask...? You do the above, you can fly a 747...
xxx
Why would I ask you to fly on all engines, if you can fly with 1 or 2 engines "out"...
Why would I ask you to put flight director and autopilot "on". I know you use them when flying.
I know you can call the check lists. It is the PNF and F/E who handle them anyway.
All these things would take an extra hour of wasted time.
xxx
With the same philosophy, I could do a F/O check in 20 minutes.
F/Os do not require 2 engine approaches...
So, all they would get is a V1 engine failure takeoff, with engine fire on same engine.
And they would complete an ILS to minimums with engine-out. Full stop.
No autopilot, no flight director.
xxx

Happy contrails

[Denti]

These have happened yet the only time stall recovery is practiced is in initial type ratings. Even then the landing configuration stall is usually limited to 5000 ft agl for goodness sake.

Interesting, we do it a bit differently over here, stall recoveries are a fix part of every sim check and training (one each every six months) and of course landing stalls are done on where they can happen, fully configured below 1000ft AGL. Sadly we dont often do high altitude stall recovery though, that could need some improvement.

[SNS3Guppy]

We do all the air maneuvers on each recurrent. Initial is 32 hours in the sim, and recurrent is eight. That includes stalls.

[Tee Emm]

. First takeoff, engine failure at some 5 knots below V1... Rejected T/O...

That brings up some interesting points of view and this is not criticism of your training method by any means.

An engine failure at five knots below V1 and thus a reject on a limiting wet runway will almost certainly result in an over-run. Probably same result on a dry runway, too. The rate of acceleration on all engines at V1 minus five knots is so fast that it is a good bet based on previous accidents the aircraft would be beyond V1 by the time you closed the thrust levers. No problem on a long runway - but should we not be training for limiting length runways where a decision to stop or continue could go either way.

When training or testing (if you like) a pilot for competency in the rejected take off procedure, the purpose of the exercise (or practice) is the correct management of braking, thrust levers, speed brake operation and subsequent actions after stopping.

To "test" the lightning fast reaction of a pilot to decide stop or continue when you are talking about a five knot decision point is open to argument. There is no shortage of accident reports and research into the danger of high speed aborts near V1 on limiting runways. In most cases the accident would have been averted if the pilot had continued the take off albeit from below V1. There is evidence that pilots should generally be "go-minded" when approaching V1 on a limiting runway. With a fire warning five knots below V1 it may be safer to continue the take off and shut down the engine once safely airborne.

Of course in theory the aircraft should be able to stop if all actions to stop are perfect at just below V1. But from what I have researched and observed during simulator training, the chances of a stuffed up aborted take off on a limiting length runway are considerably more than the chances of a crash into the trees in the go case.

If an engine failure during the take off run is required as part of the training syllabus, then I suggest 15 to 20 knots knots below V1 makes it obvious a reject is required and the instructor can then assess the crew actions rather than face the possibility of quibbling over a five knot below V1 stop or go. Keep in mind that research has also shown that with a tyre burst during the take off roll, it is generally safer to continue the take off when within 20 knots below V1 to cater for loss of braking efficiency.

As tyre bursts are more common than engine failures at V1, then giving the pilot a tyre burst at V1 minus 20 knots and opting to continue the take off, is perhaps better training than the engine failure at V1 or a smidgin below. In any case the instructor could if he so desires (the bastard..) cause an engine failure due debris ingestion just as the simulator gets airborne. But that wouldn't be nice, would it?

One exercise that should be considered during simulator recurrent training is where a landing gear (nose or main) collapses without warning on touch down or in the early part of the landing run. The FCTM and QRH for the 737 (for example) gives instructions and advice for the planned situation of partial or gear up landing - mentioning use or otherwise of speed brake, reverse thrust, fuel pumps among other things. In other words it assumes you have time to read the good stuff.

With an unplanned event such as an unexpected partial or gear up landing on touch down, the majority of crews observed in the simulator were unable to instantly recall whether or not quickly cancelling reverse, or not operating the speed brake or not, switching off the fuel pumps or not - were vital actions under the circumstances. Some pilots cannot be bothered to be familiar with the advice contained in non-normal checklists unless the items are boxed Recall Items. Ostrich head in the sand, comes to mind. In a landing gear non-normal as described above - ie no prior warning - their lack of knowledge could have serious consequences. This is why simulators are so valuable in the right hands.