B735: Why disconnect the A/T if you disconnect the A/P?

[TangoAir]

Hi,
What is the reason to disconnect the A/T if you disconnect the A/P? (That’s what the FCTM suggests!)

What does your SOP’s says about this?

[Navigator33]

Our manual doesn´t say anything about it.
But I heard they want to make it mandatory to
press speed so to keep the A/T armed in case of a G/A.
I personally do not like this procedure.

[Willit Run]

If the auto throttles are on while the auto pilot is off, you have two different computers working, usually against each other. Your brain and the airplane brains don't talk to each other, and you can get some pretty wild swings in pitch and power.
The auto throttles and auto pilot are designed to work together, how ever, auto throttles can make your job very hard if the auto pilot is off.

I don't reccomend using auto throttles with out the auto pilot. Try it, and I think you'll see the wisdom very quickly.

[Seat1APlease]

What do you do on a manual climb out? Use manual throttle and keep adjusting it or use the A/T?

I don't see any good reason not to use it, or are you just talking of speed mode at low level?

[Navigator33]

I think I know what Willit Run is trying to get at.
If you fly manual with A/T on during low speed (e.g. during a visual)
and you get too close to the speed the A/T starts powering up
which will result in a pitch up.
If I fly manually from say 15.000 what I normally do is go to
V/S first and set the speed to 210 (clean) and disconnect A/P
after that. This way I can do whatever I want with the power and speed
while still descending with idle power.
but I will have some sort of floor protection for possible underspeeding.
Same with a visual. I go to CWS and set the altitude and heading for
go around and after that disconnect the A/P. The only thing the PM
has to do is select lower speeds when I start configuring.
We have lazy skippers you know, they don´t like doing the PM MCP part (or sometimes don´t understand what you say in English but that´s a different story)

[Checkboard]

The question normally comes up for a manual approach, but with the autothrottle engaged. My last airline allowed it, my current one doesn't - so it's a fairly even bet. Boeing doesn't recommend it for approach, as it is possible for the autothrottle to increase power in the final stages of the flare unexpectedly as the autothrottle moves into alphafloor mode.

[Bullethead]

I personally believe, that on descent and approach, you either use all the automatics or none and hand fly the aeroplane. That way there is no doubt in your mind as to who's flying what. I doubt any of you would choose to operate the control column and let your co-pilot control the thrust during an approach. It's a set up for a screw up.

Regrads,
BH.

[Centaurus]

I doubt any of you would choose to operate the control column and let your co-pilot control the thrust during an approach. It's a set up for a screw up
Interestingly there are check captains/simulator instructors who insist that on a manual reversion you should fly the approach just like that above. In other words split the duties. I don't know why that is because it complicates the pilot's handling. On requesting Boeing advice on the subject, the reply was that Boeing do not have any procedures that require two handling pilots.
It never ceases to amaze me why pilots so often choose to disregard the manufacturer's recommendations in matters of aircraft handling procedures. The business of leaving the AT engaged or AT "speed" deselected in the 737while manually flying is a case in point. The FCTM is perfectly clear in it's recommendations in this regard.

[RAT 5]

I asked this related question a few years, and never did receive a satisfactory reply.
TCAS. In the B733 the QRH NNM left the A/T engaged, disengeage the A/P if appropriate. I then converted to NG and it was A/T disconnect. Why the difference? I never did find out. I do not have a currect B733 QRH, so what is the Boeing NNM procedure now?
In a TCAS climb why not disconnect A/P, increase pitch and let A/T control thrust? That seems safe. TCAS descent,same thing. I've seen, in the sim, pilots disconnect, pitch up without adding enough power - unsafe. After TCAS descent: I've seen pilots level off and forget to add power soon enough, thus with the turbine lag from idle they went slow on speed.
So 2 scenarios which were hazardous i.e. TCAS avoidance, made potentially worse by disconnecting an automatic system.
I agree that flying an approach in manual & A/T on is not ideal or easy on B737's. Other a/c types are designed to do so, so that's not apples & apples. It could be a feature of the brain that if you tell it to coordinate the lateral & vertical path only, and forget the speed because it will be taken care of automatiucally, then your scan will delete speed. If you keep all mechanics operating and coordinate both hands and both eyes, then hopefully your scan will be complete and thus safer.

[A37575]

In a TCAS climb why not disconnect A/P, increase pitch and let A/T control thrust?

For the same reason if you got a GPWS Pull up warning in IMC and you pulled up to 20 degrees immediately and waited for the autothrottle to react to the falling airspeed. Every second counts and although the TCAS is perceived as less urgent, you should still apply instant manual thrust and get going upwards. Of course a GPWS pull up warning is infinitely more serious and requires instant full bore thrust and pitch up. And don't forget the speed brake if it is extended. Leaving the autothrottle to increase thrust with falling airspeed is demonstrably too slow for safety.

[Willit Run]

Seat 1A,
I was talking about flying an approach in the speed mode. I guess I should have been a bit more clear. I do use auto throttles in climb whilst handflying because its an EPR limit and not a speed limit.

Flying an approach with the auto pilot off and auto throttles on can be quite challenging while trying to maintain a selected speed.
All I can say is, try it! if it works for you, fine, but I think you will see how unstable it can be.

[galdian]

For interest did a convert from classic to NG a couple of years ago, ran across one of the head trainers from Boeing and this question was raised.

Basically Boeing consider the 737 as their entry level jet; for a significant number of students it is their first jet and Boeing want to keep the transition as easy and simple as possible hence their general procedures reflect the "all or nothing" with regards the autopilot/autothrottle.

Bear in mind over the last 10-15(+/-) years there has been a substantial increase in self funded training Vs airline structured training - the death of legacy carriers with structured, ongoing training Vs the emergence of LCC who will employ anyone from anywhere with a variety of background training I suppose could be a fair suggestion??

In a structured airline they were always happy if "deselect" rather than "disconnect" the autothrottle was procedure knowing the reasoning behind the policy, the benefits and the "traps", would be well explained by the training department.
In other circumstances I suspect the KISS principle applies.

That's the guts of it - my apologies to Boeing if I have misunderstood anything.

On a similar note - having seen some superbly executed minimum circling approaches utilising various autopilot/autothrottle modes and selecting/deselecting as required I have no doubt the reduction in work stress adds immeasurably to flight safety.
I have no real desire to be down the back in an aircraft where the Captain decides to turn it all off and show his ability on a dark and stormy night - airline aviation (and the individuals involved) should have grown out of that decades ago.

Cheers
galdian :ok:

[Navigator33]

A bit off topic question but how many of you fly the 737 in CWS every now and then?

[bonernow]

In my previous company, on a manual approach and landing, we disconnected the AP and AT together. However, in my new company, on an approach, once stable, the AP is disconnected and the AT is then left in the 'ARM' mode. Therefore if the speed washes off the throttles will stand up and there is some degree of alpha protection.

But I would have to ask the question, "What the hell is the PM doing to allow the speed to decrease by 10 - 15 kts below vref, causing the AT to come out of 'ARM' mode and into MCP Speed?"

I can see merits for both techniques. Having the AT in the 'ARM' mode on an approach would be advisable, say in the event of an incapacitation and PF is single crewed. But with two pairs of eyes peeled on an approach, surely one pair of eyes will pick up on an adverse speed trend and make the appropriate call?

[RAT 5]

Which brings us full circle to another thread asking about the wisdom of knowing and setting the correct N1% on approach for final flap configuration and then making small changes about that datum. If that technique was followed the speed trend of which you speak would be avoided to a large degree.

[gas-chamber]

If your airline wishes to go against the Boeing FCTM or any other procedure that Boeing recommend then the regulator should insist that a statement goes in your operations manual to the effect that the airline accepts full responsibility for the changes.

Lawyers can and will take an operator and a pilot apart because they mess with procedures. It may have nothing to do with the case, but they will dig anything up if it stands an even remote chance of winning a point. Think of a large insurance claim.

Personally, I always follow the manufacturer procedure unless the person who signs my pay check puts a contrary instruction in writing in its manuals, with that statement about accepting responsibility or a statrement that the procedure over rides the FCTM or AFM etc. Over the years I have had a few arguments with check pilots for this attitude, especially those with their own pet theories. But so far no one ever fired me or even failed me on a check ride for such a stance.

Back to the automatics - use them all, or hand fly. Hand flying is not that hard if you do it often enough, but maybe not a good idea in really poor weather. Switch the autothrottle out for manual landings, just like the book says. Once the landing is assured, why would you ever need the A/T ?

Now when there is some sort of emergency not covered by the book, it is a different story. Use anything at your disposal, including four hands flying if that is what it takes. Example being jammed or seriously damaged flight controls. Boeing says that it is OK to go outside the square if you really have to.

[john_tullamarine]

If your airline wishes to go against the Boeing FCTM

Not quite on the mark. It is fair to say that the OEM knows a fair bit about designing and building planes but that doesn't necessarily extend to be the only competence with respect to operation. Particularly after a Type/Model has had a reasonable Industry exposure, the larger operators become a potent source for operational matters.

It is not at all uncommon for operators to choose to do things a bit differently to the OEM's procedures, especially other than for non-normals and emergencies. It is appropriate for such operators to seek an NTO (no technical objection) from the OEM, just to make sure that they haven't missed something hidden in the bowels of the design and certification processes. Depending on the jurisdictional requirements, it may then also be necessary to obtain local regulatory concurrence or approval for such variations.

Having seen a reasonable range of ways of doing things, my preference is to run with the OEM for simplicity and cost ... the documentation is all there and doesn't need an army of technical clerical folk to keep track of changes at both ends of the chain ...

[A37575]

QUOTE] have no real desire to be down the back in an aircraft where the Captain decides to turn it all off and show his ability on a dark and stormy night - airline aviation (and the individuals involved) should have grown out of that decades ago.

[/QUOTE]

You have hit the nail on the head albeit without realising it. The dark and stormy night circling approach took real piloting skills and fine airmanship. It proved inevitable that this type of craftmanship would fade into history as the magic and superb reliability of automation took over the pilot's job and he was reduced to a button presser. Some cynic once said "I can't fly for nuts but I can type at 80 words a minute". It is practically a MayDay situation with some captains who are so afraid of switching of the flight director until parked safely at the aero-bridge.

[bonernow]

RAT 5,

You hit the nail on the head. I was taught 63% N1 for Flap 40 and 57% N1 for Flap 30. These ball park figures DO work!! A little fine tuning around those figures for weight, wind, glide slope angle etc is all that's required once you're stable at those power settings.

[PK-KAR]

You have hit the nail on the head albeit without realising it. The dark and stormy night circling approach took real piloting skills and fine airmanship. It proved inevitable that this type of craftmanship would fade into history as the magic and superb reliability of automation took over the pilot's job and he was reduced to a button presser. Some cynic once said "I can't fly for nuts but I can type at 80 words a minute". It is practically a MayDay situation with some captains who are so afraid of switching of the flight director until parked safely at the aero-bridge.

At least over here being able to do such craftmanship is a requrement for the crew (at least in the sim sessions)... Some still do it on a regular basis and do it nicely... And those who can't handfly without an FD would be asking for extra sim sessions or flights with the checkpilots... coz they won't be able to fly into 10% of the company's routes...

PK-KAR

[galdian]

Whilst using N1 settings as a starting point may well be fine a statement along the lines of "a little fine tuning" scares the hell out of me and is indicative of a mindset for simulator work which quite often falls apart when dealing with real weather in a real aircraft with passengers on board.

I have seen the most perfect circuits etc flown by the numbers in the sim, however you start to alter a few things (weight, variable winds etc) and they'll sit there wondering why it's not working - after all they've set the numbers correctly!

Maybe its all about emphasis but these people then pass on to trainees an expectation which may work much of the time on a 747 or similar size with greater stability but could spell disaster on a light twin.
Making these guys realise it is up to THEM to push and pull AS REQUIRED and use the thrust levers AS REQUIRED to put the aircraft where it MUST BE and within the acceptable speed range, rather than where the numbers say it SHOULD BE, is an interesting experience. :ugh:

As I said maybe it's all about emphasis, maybe it's just my neck of the woods where this problem arises??
Or is it just me who sees this as a problem??