The Danger with Rapid Rotations
 

Discussion came up over coffee about the rate of rotation used by some pilots. F/O said he was consistently amazed at the fast rate of rotation of one captain in his company flying 737NG variants and couldn't understand why there had not been a tail-strike so far. He felt that a comment would not be received gracefully so merely gritted his teeth.

Some years ago I transferred to another company flying the 737-200. The chief pilot criticised my rotation rate as being far too slow (in his opinion) and warned this could compromise obstacle clearance with engine failure. I was surprised at this as no one had criticised the rotation rate before because I had stuck strictly to the one potato - two potato up to 15 degrees.

Over the next few sectors I then observed carefully the rotation rate of the crusty old chief pilot of my new company and stopwatched him consistently on 2.5 seconds from Vr call to 15 degrees. As it was a -200 then wasn't much danger of tail strike but a high speed stall at that rate of rotation was a danger particularly at Flaps 1 take off.

The CP did not believe me when I said he had been timed at 2.5 seconds from Vr call to 15 degrees up and I realised he honestly thought he was doing the job at the FCTM ideal of 3 degrees per second.

So apart from the possibility of a tail strike, how dangerous are rapid rotations? In the early years of 737-200 operation, there were several documented instances of sudden pitch up and wing drop on Flap 1 rapid rotations especially if there was traces of frost or ice. This led to Boeing advice that if ice was around it was better to use greater flap settings or improved climb speeds to give a greater stall margin and also to ensure rotation rates were as per FCTM.

I believe that some pilots are quite unaware of their fast rotation rates aalthough one would think a diligent instructor would spot their bad habit during simulator training. For all the 737 variants there is a momentrary pitch force change around 10 dgrees body angle due to the tailplane encountering ground effect and it often results in a slight hesitation at 10 degrees before normal stick force returns. Some pilots will either unconsciously or deliberately force the rotation through that area and maintain this additional pull right through to the nominal 15-18 degrees. This results in a rapid rotation before the pilot is aware of it. A rapid rotator should be picked up in the simulator because it is a habit that is hard to crack. It begs the question should a pilot who is a rapid rotator be subject to a comment by the other pilot who may consider the manoeuvre potentially dangerous? Or do you play the shrug your shoulders game and keep quiet? Comments welcome.

Currently a hot topic here in the WoT.

2 tail strikes last year (757 and 763)

All new pilots are VERY carefully monitored on rotation rates.

Very hot on it in the initial TR Sims and during line training where if you are rotating too quickly you will soon be told about it either verbally or with nudge of the column.

hi 'g' ground stall if you really fool around:}

also rotation rate is affected by the actual aircraft perf. really heavy rotate slower, really light, faster, mid weight mid perf rotate @datum rate. don't want to do all that then rotate at the rate scheduled in the afm. i.e datum rate

Well, I don't fly with crusty old Chief Pilots, but I have noticed a time or two where a new (sometimes not so new) First Officer will try to rotate too rapidly.

I have NO idea where they might get this idea, but I come from the old school, flying old 4-engine types early on, and if these were rotated too fast, especially at heavy weights, getting airborne would have been doubtful, at best.

In addition, a slightly slower rotation rate will enable the flying pilot to successfully handle an engine failure scenario, when and if that engine (especially an outboard one on a four engine type) fails at the most inopportune time during the takeoff...just prior to, or at rotation.

I watched one young guy in the sim rotate with such vigor, that during the wing engine failed just at rotation exercise, he found himself trying to climb at just under V2 at a heavy weight...and getting absolutely nowhere.

Over the years I worked on several (at least a dozen) Tail Drags/Tail Strikes and I can't remember one that occured on takeoff, all were on landing. Does any one know the ratio takeoff to landing?

I have yet to see any airplane manufacturer who suggested anything other than a reasonable rotation rate - and almost universally that is 3 degrees per second, and 6 seconds to 18 degrees. You are absolutely correct about the history of the B737s, especially the -200 series. And I know that Centaurus had written an article on this specific circumstance some time back. It's good reading.
I think that any time either pilot sees the other guy rotating like what has been described here, it would be not only a smart thing to do to bring it to his attention, it could be the best conversation of your career. It borders on being unsafe, and at the higher gross weights probably is unsafe. There would be nothing more sad than to have your friends mention the fact that you knew it was dangerous but you decided to keep it to yourself, and do so at your funeral. And, by the way, if the other guy tells you to "stuff it," I think a chat with the Chief Pilot might be in order. Sorry guys, but he asked.

Rapid Rotations during High Performance Take-offs.

To be objective in examining the manoeuvres from rotation to initial climb away for aircraft having high excess thrust you should consider the two basic segments of the rotation and throw in the effects on the human occupants.

Segment 1 would be the tranistion from ground roll to lift off.
Segment 2 would be the requirement to change the vertical direction of the aircraft from horizontal at lift off to the initial climb away angle.

Ignoring smoothness/human effects the initial rotation could be almost instantaneous to rapidly increase the wing lift to fully support the aircraft weight but one should not overlook the total rotational momentum then generated which would inevitably cause some overshoot towards a potential tail strike before adequate development of the second segment.

Having lifted off at a defined safe margin above the 1g stall you now enter the next segment which is a 1g+ manoeuver to change vertical direction through to the initial climb angle during a period when continuing increasing air speed should be allowed thus maintaining an initial margin over the small increase in the 1g+ stall.

All of these factors including human effects will have been embraced in the flight tests to determine the best rotate V for weight by TPs who expect that most pilots are trained to operate flight controls in a positive and smooth manner with some emphasis on smoothness.

I am not sure you are right Rhovsquared. Apart from the engine failure rotation the Boeing 737 FCTM advice on rotation rates does not differentiate between light weights and max gross. From observation I believe that with a light weight and thus a rapidly accelerating aircraft there is sometimes a tendency to hurry the rotation where the pilot perceives the airspeed is rapidly building up just after lift off before he is ready for it.

Unless your aircraft has some degree of intelligence built into its pitch target for rotation, instantaneously achieving the target pitch target is not necessarily going to be a good thing. Under normal circumstances the aircraft is going to start to lift off significantly before that attitude. Also, as the rotation is continued towards 35ft the aircraft is going to be climbing out of ground effect somewhat; therefore some of the further rotation post-liftoff is needed to counteract the loss of GE lift.

Now, if you *know* what the lift-off attitude is you could perhaps argue that could be attained almost instantaneously with little direct adverse effect, though there would still be overrotation concerns.

My first large jet training was on the MD80. Prior to that, I just yanked back on the turboprops I was flying and no one said anything.

The MD80 will be difficult to control on an engine failure prior to rotation if you haul it off the ground rapidly. I learned my lesson and would rather be criticised for rotating too slowly than have control issues for rotating too fast.TC

Likewise with the older models of the B707, AA717driver, and even with most models of the Lockheed TriStar, at very heavy weights.

Slower rotations (within reason) ARE better.

Apologies if someone else pointed it out before (if so I missed it during my scanning of the posts) .. quite apart from tailstrike considerations, performance data is based on the technique described in the flight manual so that technique should be the process routinely aimed for by the pilot ...

Rapid rotation can cause other problems than tailstrike. If your aircraft is fitted with a stick pusher, too high a rate can provoke the alpha dot term to generate a push. Most modern Stall Protection Computers ramp in the alpha dot term a couple of seconds after liftoff for this very reason.

AFAIK all the Boeing numbers are based on a rotation rate of 3° per second. So to get to 15° should take about five seconds.

Mad Scientist

I must take more care with the words!

No way was I proposing a technique for rapid rotation through about 15 degrees to the lift off alpha - just that it would be of little consequence if you did just that other than for a tendency to overshoot too early into the airborne segment. It would certainly startle those aboard but would get the wings doing their thing promptly.

Much better to aim for a constant rate of rotation through the transistion from ground to air borne knowing that when airborne you then start to pull some extra loading (1g+) to establish your initial climb angle whilst expertly flaring off the rotation to reach the initial climb attitude coincident with the initial climb IAS. Now that takes a bit of type experience and some 'stick and rudder' finesse to do it all smoothly, all of which tends to sort out the good pilots from the not so good!

Then there will be the hot day and a weighty heavy which will be reluctant to go flying. This is where experience can hone the instincts of the pilots in a determination of the optimum technique for rotation to lift off when the margins are small. For most types a slow rotation to lift off will often be consistent with progressively replacing rolling wheel drag with increasing wing drag throughout the rotation to achieve maximum acceleration. There will be precious little rotation to be concerned with after the lift off. Runway length will be a major factor and the lessons learned from rotations made too early are legendary.

A37575, I appreciate your feed back I do believe that on the 73' and later Boeings the Artificial Feel systems helps to minimise the effect on acceleration on rot. rate as well as CoG effect.

I definitely don't want to tell working line pilots what to do, but it could also be a Datum Rate meaning a Mid-perf test rate that protects against the stall or hi 'g' throughout the extrema of the performance (AEO and OEI) range with out comprimising distance.

To look at the raw physics of it (as well as that info being in a well known TP's book;) If you ingore flight path momentum for a moment, as it is high in both cases i.e Mv or mV,. and bring it down to the fundemental forces F=MA
since A is changing the F= Lift is changing as V^2, because the plane is accelerating fron Vr to Vlof through the screen at V2 or V3 the rate of change of the lift force must be met with the appropriate rate of change in AoA(incidence).

My statement does assume there is no abuse in rotation attitude or speed; edited to add rate of change of AoA

I'm sure 411A rotates at the exact rate the performance calls for i.e if 2.6575 deg/sec was called for that's what he gives:ok:

In the two aircraft types I have flown, 727 and 737-200, the recommended rotation rate was 3° per second. The 727 had tailstrike potential but on the 737-200, it was much less likely.

I was reading an accident brief the other day(for a DC-9-10) and found this interesting quote...


"Normal DC-9 Series 10 operating procedures advocate an approximate 6° per second rotation rate. The 6° per second rate of rotation can result in higher than desirable angle of attack as the airplane becomes airborne. The Safety Board believes that a 3° per second rotation rate of the DC-9 Series 10 airplane should be specified when icing conditions exist. The same procedure was applied to Boeing 737 airplanes after a number of them experienced “wing roll-off”(partial stall) at liftoff in icing conditions."

I have never seen this mentioned anywhere else. Is it still a 6° rotation rate on the short DC-9's?

Anyone have old 737 manuals(on the net perhaps) that mentions the 6° rotation rate?

I looked at the Air Florida report but nothing is mentioned about it.

6° per second? You have got to be joking. The 717 (shortish DC-9) FCOM recommends 2.5° per second and that feels too fast.

Understatement of the century. :eek:

Even 3° per second is too much on a long 737 in gusty wind conditions, that tail gets very close to the ground.

From the 2011 737NG FCTM:

I flew with a Captain back when I started on the 737-800 first and he used a really fast rotation to about 8° then leave it settle for a second before launching it up again to 15°. He used to fly the -500 for another airline before but I used to cringe every time he'd rotate and wait for a thud:eek:

I've never understood some people's obsession with trying to snatch the aircraft off the ground, all engines running. If you get an engine failure, you're going to be rotating a bit slower to a lower pitch attitude anyway, so it doesn't compromise performance noticeably. Yanking the thing into the air does make me squirm in the seat, though!

This accident involved an ABEX DC-9 in PHL many yaers ago in freezing drizzle conditions without being de-iced. Aside from rapid roll reversals, their power dropped off significantly due to turbulent airflow into the nearby engine inlets(during certification MacDac had seen this stall/power loss combination when doing high altitude stalls which happen at a lower AoA). Similar happened to TWA and Ozark DC-9's in earlier incidents. Apparently longer body DC-9's are not vulnerable to this as the turbulent airflow goes above the tail.

It is interesting to note that for baby DC-9's, the slower rotation rate recommended by the NTSB and the procedural changes concerning rotation rate made by ABEX only applied to icing conditions.

Slow Vs fast rotation rates.
 

I agree with the 3 degree rotation rate. If you are not going to do 3 degrees per second then slower is better than faster. I have often shown my Co-pilots how close some A/C tails come to touching the runway on take-off by making them observe actual take-offs at various runways while on duty. A slow rotation rate also gives the A/C time to get to correct flying speed if the take-off speeds have not been correctly worked out (within reason).

Bad science! It's a cost/benefit analysis between lower obstacle clearance (1% of take-offs are first or second segment obstacle limited) VS running the speed on one engine too low (100% of take-offs with an engine cut at V1 are at risk of this problem) :hmm:

If your rotation is too low - energy is still being added from the engines, so if you see a tree in the window, you have that energy still available to pullup.

If your rotation is too fast - well, the closest a normal flight EVER gets to the stall is at rotation, and stall, flip, crash, burn, die is the classic. :uhoh:

If you are TWO engines, then you have so much extra power over the ONE engine case the performance is planned on, it's not a problem.

If you lose an engine, you will have less pitch-up inducing thrust, which will lower your rotation rate in any case.

Another very important point. Never "double rotate" i.e. rotate to say 8 degrees and stop, then rotate to 15 degrees. When at 8 deg, the aircraft may leave the ground but will not climb more than a few feet. So there you are flying just a few feet off the runway then you rotate a second time, driving the tail into the ground. With a constant 3 degrees per second, as the angle increases the rate of climb also increases. This allows the tail to descend towards the runway at about the same rate as the aircraft is ascending. A win-win situation.
Know your airplane.

I have to admit that I did both in the 727. A continuous 3° per second rotation to 15° and sometimes a hesitation at about 10° pitch later in my timeperiod flying them, but the mains were still on the ground for the momentary hesitation. I don't believe that Boeing says anything about a slight hesitation being the way to do it.

Constructive criticisms are welcome.

If so critical.. do any jets have an auto rotate feature or rotation limiting features?

I have in other recent threads been advocated strongly that people look out the window when completing the final part of the approach. By that I mean stop looking at the instruments (except for airspeed) and maintain lateral and vertical path by looking at the touchdown point at below 300 feet latest (weather dependent of course).

Apparently, the need to look out the window needs to be emphasized when it comes to takeoff as well.

When PM calls V1, I have a confirming look myself and thereafter only look out the window until established in the climb, completing the takeoff by going back to the instruments and adjusting pitch to maintain V2+20.

By looking out you have a complete feel for how the aircraft is reacting to the rotation and adjustment is then much easier. Plus it makes spotting drift and maintaining centerline much easier as well (from lining up behind departing aircrafts apparently a discipline that few care about in these days). The feel for the right rate is much easier to spot in a real 3 dimensional world, than on a 2D instrument.

To look out the window you have to be seated properly at the designed eye reference position. All too many are seated completely wrong, too low and too leaned back (but I guess they think this increases the coolness factor :rolleyes: ).

Exactly!
Looking out during last few hundred feet and during rotation is essential.
Rotation should be a continuous increase of pitch(just as the flare)to be inside the flight director/15deg at V2+15.If you end up with an airspeed below you over rotated, if above you under rotated.
3deg /sec too complex for me.:E

Years ago I saw a film of a live demo by Boeing on the 727 in which they set up a 35' stick, and a camera at the DER and did a series of t/o's. Using all engines they did under rotation, over rotation, too slow a rotation, and too fast a rotation. On each of these 4 t/o's with all engines running they were under the 35' mark as they launched. On the last t/o, they pulled the center engine to idle at V1 and continued the t/o. With a normal rotation, they exceeded the 35' mark as they launched. This was a perfect illustration to a simple guy like me of the importance of proper rotation technique.

I agree in theory with if you are at or around V2 + 20 having a done a smooth continous rotation then you are doing it at the correct rate....although when you're very light I find this not to be the case.

Just a quick one - I believe the above poster flies 737ng where the FDs command V2+20 after rotation not 15..

On an allied subject and that is FD reading after VR. Assuming engine failure just after VI where the aircraft has not yet reached VR on one engine. Providing the rate of rotation is correct it is my understanding that the FD will give the correct angle to maintain V2. If the rotation rate on one engine is slow then it is probable the FD will not give V2 but a different angle in order to maintain the airspeed at some figure well beyond V2.
Question therefore: If the engine failed while on the runway at V1 and the take off is continued, and the pilot is slow to commence rotation with commensurate increase in airspeed above the scheduled V2, does the FD command V2 or some other speed and what speed is that. It is known that the FD will adjust its angle to maintain the speed at which the engine failure occured after airborne - but how does the FD change with slow rotation rate if at all?

Depending on type of aircraft it is not advisable that you fully look out of the window until end of rotation. On widebodies, your margin between not climbing and striking your tail is so narrow that you have no chance to guess your pitch by looking out of the window - you basically only see your dashboard as a line witout horizontal reference of the wings or other parts of the aircraft.

I recommend a "fading in", i.e. let your scan change from fully looking out to fully looking in in a gradual sense, like you do it during landing when approaching to minimum.

About the rotation rate:
It is widely accepted that you fly 3°/sec. On bigger aircraft and during OEI, it is slightly lower. I also heard from old school DC-9/MD-80 pilots that you should rotate fast, because if you don't get your climb attitude fast, you will not climb.

Dani: Amen!

With the "look out the window during the whole rotation" brigade I have always wondered just exactly they thought that they were looking at during a low visibility take-off!

99% of the jets of 737-800 size and larger I watch never go close during liftoff. 5 seconds from 0° to 15° ish? It just doesn't happen. In my little MD machine, ours is recommended at 2.5° per second, and if done at that rate, the rotation feels positively horrible. The only thing that saves us from scraping the rear end is...we don't have one.

I suspect only because of the very inefficient wing. 20° nose up with a flight path of 12°. Looks impressive though.

Boeing suggests a rotation rate of 2° to 3° per second on the 737 depending on variant. Longer airplane variants should be close to 2°/second, smaller close to 3°/second. The slower rate is probably a good guideline for the 900ER and 800 variant, 700 should be somewhere in the middle and 600s at the faster end of the range. Target attitude is usually 12° but depends on conditions. Usually take offs are done using both derated thrust and assumed temperature which leads to low power take offs with a low pitch attitude. Unreduced take offs however, especially on light weight, are closer to 20° pitch.

There is not one rate that suits all airplanes, simply check your FCOM use the correct technique for the type in question. A blanket 3°/sec won't work.

That's because you are most likely seated too low.

With the recommended seating position to obtain optimum eye reference position you should have "sight along the upper surface of the glare shield with a small amount of the airplane nose structure visible" (according to the NG FCOM, most likely your aircraft will have a similar recommendation).

So you should have the nose visible and in that case it is very easy to judge the angle (and rate of change) compared either to horizon or other ground reference - far easier than on a 2D instrument, depth perception does wonders in judging speeds, distances and accelerations.

When we learned to fly a SEP straight and level, we looked at nose versus the horizon - basic flying. Obviously there are situations where we have to use instruments as primary reference when flying airliners, such as when the nose is pointed 20 degs upwards and it's no longer possible to see the horizon, but that doesn't mean we have to discard basic flying techniques all together, in particular not in situations where they are superior.

Low vis takeoffs are another story and just because we use a special technique, doesn't mean we have to apply that to the 95% other takeoffs we make - we don't do auto lands all the time either, do we?

Besides, described technique can be used when the weather misty too, you don't need 10+ km visibility to have a reference point on ground when at such a low altitude. If the cloud cover is low, switch to instruments earlier - common sense?

Nope, as all modern jets we do have those nice little ball thingies giving you the exactly correct eye height (in my case: Airbus).

Look (litteraly), a human brain is simpy not able to judge attitude to the last degree. In a long wide body jet aircraft, the difference between 10 and 11° is in the air or tail in the tarmac. Look at the skidmarks in Melbourne...

btw you don't see the nose if you are seated correctly in an Airbus, and I think to remember the same is valid for most other jets - except the Concorde with nose not lowered.

May I ask what type you're flying?

Even during landing we have to check our primary flight display. You simply cannot make a landing purely by feel and luck.

Dani

Dani,
I am flying 737. Above I wrote an exact quote from the 737NG FCOM, to repeat:
"sight along the upper surface of the glare shield with a small amount of the airplane nose structure visible".

If you don't see the actual nose of the Airbus, it won't make a difference. My point is that if you sit high enough you should still be able to see the horizon and not have it "hidden" below the glare shield. The upper surface of the glare shield should be just as good a reference as the nose structure.
On a day with low weight and a long runway (hence heavy de-rate and improved climb), the V2+20 pitch is often only 10-12 degs, from this pitch it is still possible to see the horizon when sitting in the correct position.

I seriously doubt that there is a 1 deg margin on any longer jets (wide body or not), unless someone screwed the rotation up, or like Melbourne used wrong data (which makes it a pretty bad example).

Hence, it shouldn't be necessary to judge to the last degree, if that's necessary something was done wrong (like the "too fast initial rotation, pause" technique). You should be judging a rate instead, that you don't need to make a pause since the aircraft is safely in the air long before passing 10-11 degs. That is my point in case - you will be much better at judging a rate looking outside than on a 2D instrument.

I am flabbergasted and speechless by your comment about landing. Just to make sure, you are saying that a safe landing accomplished by looking out the window is due to "luck"?? :ooh:

Not by pure luck, but by feel and luck, you twisted my words.

Let me assure you: On a wide body it is one degree that makes a difference. An A340-600 has something like 9.5° tail strike clearance but needs 12.5° initial lift off pitch attitude - make the maths. A widebody you cannot fly just by looking out of the window. And responsible TRTOs on medium range aircraft teach their pilots to fly them the same way like their bigger brothers - to get a common mindset.

Dani
(A340-600 rated)

Again, I can't see what relevance the width of the body has to do with tail clearance. What is it that you would like to say with that? What is it that in your opinion makes a (oh lord) wide body different from other aircraft?

Actually this is what Boeing has to say about it:
I don't have a wide body Boeing manual at hand. But normally, as you say, they maintain a uniform philosophy across their range of aircrafts. Exactly my point again. Focus on getting the rate right, and the performance calculations takes care of the rest of your worries. The topic is "the danger with rapid rotations", with the right rate, the rotation won't be rapid. I am giving my point of view on how to achieve the right rate.

You are missing that the wheels unstick at a considerable lower pitch than 12.5 degs (hopefully, otherwise you are missing 3 degs on every takeoff :E).

To take care of the landing as well, here is what Boeing has to say about that:
They don't recommend looking at the PFD during the landing. Again, I am sure e.g. a 777 manual will have the same wording.

The techniques above are applicable to all 737 models, including the 900ER which have a Tail Strike Pitch Attitude of 10 degs during takeoff and 8 degs for landing.

If you can provide a reference from the your Airbus manual with a recommendation to look at the PFD during rotation and flare, all I can say is strange aircrafts, strange ideas! However, I doubt you can...?