"Push" recoveries

[BOAC]

Slightly off topic but I should add a caveat to FullWing's post

I made a grab for the control column but was too late

- in case anyone should get the wrong idea, it is not a good idea to attempt to correct a potential heavy landing by pulling back on the c/c, as the wheels will merely impact even harder into 'terra firma', raising the 'g' reading. The application of a small amount of thrust will normally help to cushion such, and for the more accomplished, easing FORWARD on the stick can help (but BE SURE or don't try that! )

[FullWings]

BOAC,

I think the subject of intervention on the controls is somewhat contentious and depends entirely on the individual circumstances.

I think that if you are heading towards the ground at 800fpm and are more than a second or two from 'touching' down, then any input to increase the AoA of the wings and start to reduce the vertical speed component is a good thing.

I also propose that the the old 'drive the wheels into the ground' chestnut is mostly a 'red herring'. (Excuse my mixed metaphors!)

The wheels are not that far from the C.G., about which you are attempting to rotate the aircraft. Look at how misloading can tip an aircraft onto it's tail for proof of that. If you work out the relative motion of the wheel units for a given change in body angle, it's pretty small.

At the end of the day it's about how the mass of the aircraft is decelerated as the suspension compresses beneath it - there is a fixed amount of oleo travel and if you use all this up your landing will become very much harder.

If the travel is, say, 5ft then you wish the rate of descent of everything else attached to the upper end to reduce to zero at 4'11 3/4" travel. Any reduction in V/S prior to touchdown will help this be achieved.

Extra thrust? Nice if you can get it but with today's big fans you'll probably get a good whoosh of power after impact! Also, if you resolve the forces, given normal touchdown attitudes, the amount of force applied to the mass of the aircraft normal to the runway surface is totally inadequate to produce any significant change in the rate of descent. I think this other 'old chestnut' comes from the days of propellors where the wash over the wing produced an increase of lift as well as thrust.

To get a jet aircraft to radically change the direction it's going in a short space of time in you have to change the AoA of the wing (thrust vectoring excepted).

[Maximum]

I know we're off topic, but I just have to respond to your comments FullWings.

I know from experience that I can cushion an impending hard touchdown at the last moment only by an application of thrust - if I pull back on the controls, the sink rate stays the same and the landing ends up harder.

Also, quite definitely, in experienced hands a 'greaser' can be achieved time and time again: 1.) by having the correct r.o.d. in the last hundred feet or so and 2.) after the flare and just before the mainwheels touch, pushing forward slightly. (Non standard I know).

My own experience of this is absolutely in line with BOAC's.

[FullWings]

Interesting. If you pull back on the controls, the sink rate stays the same. Does this work in other phases of flight? Maybe we should revisit the actions required for a go-around or hard GPWS?

How do you know you weren't going to get a good landing anyway the time you 'pushed forward' on the controls? I think a 'greaser' has more to do with your ROD during the touchdown versus auto-speedbrake deployment but I digress...

I'm just pointing out that if you actually analyse the physics of what you think you are doing, it might be somewhat different to the common perception...

Try this while you are on approach:

Note the rate of descent. Add thrust (keeping the attitude the same). Note rate of descent a few seconds later.

And this:

Note the rate of descent. Pull back on the yoke, keeping the thrust constant. Note rate of descent a few seconds later.

Notice any difference?

{Apologies for the creeping thread, all you lurkers!}

[411A]

Maximum,

Yes, the 'Boeing push' generally works quite nicely, if done properly...from the very early models of the 707, to present day types.

A couple of older types, it didn't work so well....Lockheed TriStar, for example, as a few (more than a few) found out.

[FullWings]

I've just re-read this page and think that we might be talking at slight cross purposes about the whole 'pushing' thing.

I think BOAC, Maximum & 411A are describing a landing where the rate of descent has already been reduced to a low value and the aircraft is settling into the 'ground effect'. A slight forward movement could possibly do something strange with the tailplane/mainplane combination.

I'm talking about an unchecked rate-of-descent which will likely cause distress/damage to the aircraft and passengers. Pushing forward in this situation will do nothing to reduce the energy that has to be absorbed by the landing gear/airframe. For an extreme example see the thread on the Britannia crash at Girona.

I would strongly caution any pilot about using non-documented handling methods unless they understand exactly what they (and the airframe) are doing. If it was that good, Boeing (and others) would be recommending it. That's not to say it doesn't work most of the time, just think about what you would tell the investigating authorities when it goes wrong...

[Maximum]

FullWings, of course I'm totally aware of what happens if you pull back on the stick in a descent etc.

But the point is, we are talking about circumstances within, let's say for arguments sake, forty feet of the ground with a rapidly sinking aeroplane.

So the crucial factor you are leaving out is time. Time available to arrest the sink. If we pull back on the stick, given time, yes, it would have an effect. But before that happens, we've already touched down - harder than desired in our example.

I say again, from experience I know that the application of thrust has a much more immediate effect. (And yes it does pitch the nose up slightly on underwing engined types). So perhaps we're both really talking about the same thing here - an increase in AOA and a thrust increase - which would make sense. However, in piloting terms, I'm sure it's best thought of this close to the ground as an immediate increase in thrust that's needed.

As for pushing forward for a 'greaser', as 411A verifies, the 'Boeing push' is a well known phenomena - haven't you felt this yourself? And I know from analysing my own control inputs that it definitely works (and can also be disastrous in the wrong hands).

..............break break..............*grin*

Fullwings, I think we both posted at the same time there as I hadn\'t seen your last post.

Yes, I agree entirely, pushing forward will in no way arrest or save an unacceptably high rate of descent.

Extra thrust is what you need!

[FullWings]

Hi Maximum,

Yes, I think we agree on most of this.

One small point, which I think you are starting to realise:

I say again, from experience I know that the application of thrust has a much more immediate effect. (And yes it does pitch the nose up slightly on underwing engined types). So perhaps we're both really talking about the same thing here - an increase in AOA and a thrust increase - which would make sense. However, in piloting terms, I'm sure it's best thought of this close to the ground as an immediate increase in thrust that's needed.

If you read what you have written it actually says that it's the change in AoA caused by underslung engines: a secondary effect if you like. You have pitched up without elevator input. If you fly an aircraft without underslung engines, this technique has little or no effect. Also true of FBW types which take out the thrust/pitch couple (Airbus & B777 - what I fly now, so no I don't feel these effects ).

I still put forward that the only way to reduce the rate of descent significantly near the ground in a jet is to increase the AoA (somehow).

You will eventually get more lift from the wing from the increasing IAS, as the airframe accelerates forwards but this will take quite some time to happen. Much longer than the effects of pulling back on the stick.

I'm not saying that adding thrust is a bad thing but it isn't what is providing most of the reduction in descent rate. It just appears that way.

It's basic Newtonian mechanics. You have a large mass of aeroplane moving towards the ground in the vertical plane. To reduce this rate of descent you must apply a similarly large force over a period of time in the opposite direction. Think about where this comes from...

Safe flying all

[BOAC]

All the above points are vaild, but a late 'rotation' or 'grab' WILL increase the INSTANTANEOUS 'g' at main-wheel touch, and could well put the QAR reading into the 'coffee-no-biscuits' area. Of course a pitch change at the RIGHT time is better, but we are talking about when it is 'too late'. Bringing Gerona into this is a REAL red herring. I can assure you, that having rescued a few co-pilots' potential 'heavies', 'grabbing' late is NOT a good option. All I am saying is a warning that a late intervention can often make things worse.

I 'co'd' for a very experienced Captain a few years ago who ALWAYS had the most 'horrendous' descent rate all the way to landing - such that I had the leather gum protector in my mouth - and he ALWAYS greased it on I STILL cannot work out how he did it.

[FullWings]

I'll have to get my apologies in first for hijacking this thread - it's a shame we can't branch off this section into 'Tech Log'.

I find this sort of discussion fascinating because normally pilots are quite technical in their outlook, i.e. knowing the Flying Manual, checking the weight and balance to 1kg, setting Vref to within 1kt, etc. Then you start discussing landing technique and you end up with a sort of PMT: "It just does, OK!!!"

Being of an inquisitive mind, I like to understand the reasoning/logic/physics behind what we do. Not that it seems to make me any better at it but I like to anyway.

Here's a multiple choice question (tech quiz, say):

You are approaching the ground at a higher-than-optimum rate-of-descent. Do you:

a) Pull back.
b) Push forward.
c) Sit on your hands.

I hear what several of you are saying: If you try and save a bad landing by pulling back at the last minute, it makes it worse.

This is interesting from a scientific point of view because you have only done one landing, whereas to make a comparison you would have to have done two from exactly the same starting point. I suppose I am trying to say, HOW do you know it made the landing better/worse if there is nothing to COMPARE it with? WHY should it make it worse? Let's have some sines and cosines and equations of motion...

We all know every landing is different, so it doesn't wash to say "But we did XXX at YYY and thumped it in so..."

There are a lot of 'old wives tales' floating around in aviation and it's very easy to start believing in them. Especially if your Flying Instructor (God + 1) told them to you.

People will generally believe what you tell them. Especially if things seem to work in a way consistent (on the surface) with the explanation. I reckon you could teach someone that the Rudder was the primary means of turning the aircraft and the ailerons were there to stop you wobbling all over the sky as you turned. The theory is mostly consistent with the practice but we all know it's not quite correct.

Maybe one of us should try 10 landings in the sim from a frozen position and see what happens with different control applications. This might produce some results but again, sim manufacturers are notorious for slack programming at the edges of the flight envelope.

[Maximum]

Terrible case of thread hijack this, but anyway.

What about 4.) Add thrust.

Just very quickly, without thinking about it to much, I think part of the reasoning is in the fact we have a high sink rate - perhaps because we've let the speed wash off to below Vref. We're now well on the back of the drag curve, so pulling back simply increases the drag and we slam it in.

Also, I know from experience that if you pull back at just the wrong moment you do rotate the gear harder onto the runway. It may be a small moment, but it certainly has an effect.

And thirdly, as I mentioned before, there just isn't enough time with a high sink rate and a late pull for the aircraft to change it's vector into the tarmac.

Proof of this is also seen in the case where you have a high sink rate but a margin of speed over Vref. In this case the landing can be saved with a pull (as long as it's not too late) as the lift increases rapidly with increase in AOA due to the excess speed. The ground cushion is also more effective in this case - think of the float one gets with a flapless landing.

All the above applies to certain jet transport types, not to light aircraft with so much less inertia.

Sorry there's no equations - not my strong point!

[FullWings]

Add thrust. Hmmm.

Firstly, I would like to state that I too think it would be beneficial but the question is by how much? Well, better to give it a 'handful' than do nothing...

If you get so slow that increasing the AoA on the wing doesn't give you any more lift, there is a word to describe the situation. Stalled! So at that point, you are doing a 'stall recovery' - full power, etc. When I fly something like a Piper Cub, I try and stall it a few inches off the ground. I definitely do NOT do this in a jet transport. Not only would you run out of runway but the tail would probably hit the ground.

I'm thinking of force resolution. You need somehow to reduce the vertical component of the aircraft's motion so it doesn't hit the ground so hard.

If we examine the act of adding more thrust more closely it gets interesting. Take the B777 (just because I have some of the data next to me). Go into the 'flight with unreliable airspeed' section and look up the landing config. body angle at a normal landing weight: -0.3 degrees, i.e. slightly nose down. This is almost identical to the angle when sat on the ground, so the extra thrust has a direct effect on the vertical motion of the airframe of approximately...nothing! I would assume that other jets are not that far away from this in terms of body angle.

So, we are now looking at secondary effects. Underslung engines will give an 'uncommanded' pitch-up moment with increasing power. (Not on the 777 as the FBW deliberately counteracts this.)

We are left with trying to increase the airspeed to get more lift out of the wings. Not only does this take time (you are in a high-drag configuration) but the percentage increase in lift for a 5-10kt change in IAS is quite small because of the higher approach speeds of jet aircraft.

{Explanation: I am coming in to land at 40kts in my 'cub. I apply some more power to pick the speed up to 45kts. The lift from the wing (all else being equal) has increased by (45/40)^2 = 26%. I now do an approach in the 777 at 140Kts. An increase of 5kts IAS will give me (145/140)^2 = 3.5% more lift. Not a lot, really and certainly nowhere near enough to affect the outcome.}

Where does that lead us? I still remain skeptical about anything other than pulling back on the controls but remain to be convinced...

[BOAC]

I´m asking the mod for this forum to split this when time available as it is definitely going ´thread creepy´

landing config. body angle at a normal landing weight: -0.3 degrees, i.e. slightly nose down.

FW - are we getting attitudes confused here - is that the ´landing attitude´or the ´approach attitude´? If the former, I would expect loads of nosewheel first touchdowns? I would venture to suggest that most a/c land in a nose high attitude - also what is the engine mounting angle?

My MAIN point is to try to discourage last minute ´grabs´such as you describe. Having sat through endless ´circuits and BUMPS´in aircraft from 152 through Jet Provost to just ´BUMPS´ in Lightnings and 737s, I can assure you that such ´grabs´are not helpful in all but the first two types. That is most definitely NOT an OWT.

If you have any doubts about the ´physics´ of rotational kinetic energy, take one of your most expensive and fragile wedding dinner plates, rest it on a suitable pivot (NB Zero descent rate), and then whack one side of it downwards onto a concrete floor.

To guarantee this thread going to JetBlast what about " you are in a lift which is falling down the shaft. Just before the lift impacts with the bottom of the shaft, you manage to spring upwards at the same speed as the lift is falling." Discuss

[flapsforty]

Always welcome BOAC!
I'm sure mr Draper would have a considered answer to your hypothetical.

[Flight Detent]

Hi all,

I'm confused again, we are talking about a 'hard' landing here are we not, otherwise the best suggestion I have to fix a 'heavy' landing is to go out yonder and dump some fuel!

Cheers

PS - the best fix for landing a -200 series B727 was to push slightly just prior to touch, a little thrust always helped if the descent rate was a bit large!
As I understand it, the -200 B727 was one of the most difficult to consistently land with only the required amount of 'bump' at touchdown.

[FullWings]

Thank you Mr. Flaps, it was starting to get embarrassing.

Discussion mode re-enabled.

Part I:

FW - are we getting attitudes confused here - is that the ´landing attitude´or the ´approach attitude´? If the former, I would expect loads of nosewheel first touchdowns? I would venture to suggest that most a/c land in a nose high attitude - also what is the engine mounting angle?

Well, I suppose if you wanted to be really pedantic, you would say neither! As I understand it (and am ready to be corrected) you have 'Approach Configuration' which will normally involve a bit of flap but no gear and 'Landing Configuration' which involves the gear and all the flap you are planning on using. 'Landing Attitude', I would suggest, would be the final pitch achieved during the flare, just before touchdown. Subtle differences I know but you were right to point in that direction.

Looking at the manuals again, I see on the 777 that a 'Standard Landing' involves initiating the flare at c.20R and raising the nose by one to two degrees. At the moment of touchdown the angle is c. four degs. Looking at the aircraft sitting on the ground and walking round the engines, comparing the jet pipe to the ground, I can't actually see any evidence of them being mounted other than pointing straight back, although I admit I don't know how good I am at judging these sort of things.

Let's do some maths.

Come back everyone!

Now, before we begin, let's list the assumptions:

1) 'Thrust Angle Relative to Ground'. OK, four degrees touchdown attitude rounded up to 5 in total.

2) We use the most powerful variant of the 777 that I fly (Trent 895).

3) The engines spool up instantaneously to G/A thrust. (190,000lbs.)

4) Rate of descent at 25R is 700fpm - consistent with a landing weight of c.180T on a 3 deg. approach.

5) The airframe immediately assumes the touchdown pitch attitude when commencing the flare.

I think these are pretty conservative assumptions and all err on the plus side (as far as the effects of thrust on landing go...)

Resolving forces in the vertical plane...

Component of thrust (Kg) = Sin(5 degs) * 190,000 / 2.2
=> 7,500 == 7.5T

From Newton's 2nd Law, F=ma...

75 (KN) = 180,000 (Kg) * a
=> a = 0.42ms^-2 or around 1 & a third feet per second per second.

At that rate it would take a little under four seconds of FULL POWER, just to reduce the descent speed from 700fpm to 400fpm. There is an effect but it's not as great as we might have hoped. (OK, so if it was a 'normal' landing, being at the attitude described, you should have experienced a 'normal' touchdown but I'm just using these figures to show what real effect the thrust is actually having on the rate of descent.)

By raising the nose with the elevator, you can probably produce forces an order of magnitude greater in a much shorter timespan.

In real life, the engines would take some seconds to spool up to that level and the attitude would be somewhat lower to start with. Also, we are describing a twin with a pretty good power-to-weight ratio. Finally, would you be ready (mentally) to use TO/GA power at this point?

Part II to follow after the interval

[FullWings]

If you have any doubts about the ´physics´ of rotational kinetic energy, take one of your most expensive and fragile wedding dinner plates, rest it on a suitable pivot (NB Zero descent rate), and then whack one side of it downwards onto a concrete floor.

We didn't get any plates at our wedding but the hotel did have carpet.

Take a model of an aircraft and do the same to it. Oh, the tail has hit the floor.

As I ventured in a previous posting the main gear on most passenger jets is attached fairly close to the C.G. (Otherwise you'd never get the buggers off the ground.) Using 'back of the fag packet' geometry, if the main gear was 10 feet behind the CofG and you pitched up an extra five degrees (actually enough to get a tailstrike on the 777), then the wheels would move about 11 inches (and not all of that in the vertical plane). Compared with everything else that is going on, I don't think it's going to make a huge difference to the smoothness of the landing. This example is for a 200' long aircraft - on smaller things like the 737 the effect would be much less.

" you are in a lift which is falling down the shaft. Just before the lift impacts with the bottom of the shaft, you manage to spring upwards at the same speed as the lift is falling." Discuss

How could I refuse such an invitation?

Well, assuming you are using your legs to propel you away from the lift (accelerating it further in a downwards direction as you do so), I don't see much point in doing it. Why? Because the acceleration you are giving yourself to reduce your velocity to zero w.r.t. the lift shaft is pretty much the same as you would experience when you use those same superhero legs to take the impact of the lift crashing to a stop...

[Wino]

You are forgetting one more thing about pushing vs. Pulling.

The tail plane exerts an enourmous DOWNWARD vector of lift on the aircraft (on an A320 somewhere around 40,000 lbs of force are applied DOWN on the tail to keep the nose in a steady state) That 40,000 lbs of downward force is required to be overcome by the wings as lift in addition to the weight of the aircraft inorder to lift the whole kit and kaboodle off the ground.

When you "push over" you instantly lighten the aircraft by some percentage (rather large) of that 40,000lbs. That will have a momentary lifting effect on the aircraft before the reducing angle of attack overcomes it and the aircraft starts down hill. It is quite noticable.

There are other things guys don't notice in every day flight either. You MASH the rudder for example and the aircraft will roll the otherway first (a few degrees) before the yaw developes and the aircraft rolls in the direction of the rudder.

Cheers
Wino

[safetypee]

Wino I suspect that you and other contributors to this thread are seeking a generic answer or are being too specific about an individual aircraft type. An aircraft's response to pitch control input will vary with aircraft type (and configuration) – the short period motion. Also, so will the effect of thrust, angle, amount, engine response time, etc.

For those who have used several techniques across different types I suspect that the nose down pitchers, at best, will stop the nose up pitching motion (rotation = zero), and thus if the residual attitude is less than that for tail strike a satisfactory landing will be made. This of course is speed dependent, an attitude less than for a tail strike at low speed may still result in a strike after oleo compression if there is reduced lift due to slow speed.

For the thrusters, this can be a powerful effect, not only from vectored thrust, but also ‘recirculated’ lift for wing mounted engines, and the lift from any speed increase. The effectiveness of these varying by aircraft type.

And then there is ground effect!

The nearest generic procedure that I have seen is a combination of sharp pull, check, and immediately remove what you put in (push), with a touch of thrust. Then for a co-pilots landing, a firm hand preventing further rearward stick movement with a thrust increase.

[FullWings]

The tail plane exerts an enourmous DOWNWARD vector of lift on the aircraft (on an A320 somewhere around 40,000 lbs of force are applied DOWN on the tail to keep the nose in a steady state) That 40,000 lbs of downward force is required to be overcome by the wings as lift in addition to the weight of the aircraft inorder to lift the whole kit and kaboodle off the ground.

When you "push over" you instantly lighten the aircraft by some percentage (rather large) of that 40,000lbs. That will have a momentary lifting effect on the aircraft before the reducing angle of attack overcomes it and the aircraft starts down hill. It is quite noticable.

I don't have any figures to hand but a downforce of 40,000lbs on the tail sounds rather a lot for an aircraft of only 93,000lbs empty weight? Is the aircraft so out of balance on the final approach it needs this huge corrective input?

Also, if these figures are true and you are reducing this amount by a rather large percentage the pitch-down couple is going to be huge

Help! We need some aerodynamicists!

I like safetypee's response best:

The nearest generic procedure that I have seen is a combination of sharp pull, check, and immediately remove what you put in (push), with a touch of thrust. Then for a co-pilots landing, a firm hand preventing further rearward stick movement with a thrust increase.

I wonder what would happen if you omitted the 'push' from that sequence of events? Probably not much in terms of the firmness of the landing but maybe you are lessening the chances of a tailstrike by reducing/holding the attitude just prior to touchdown. Maybe that is what this was all about to begin with?