Every primary student is taught that power controls altitude and pitch controls airspeed — or was that power controls airspeed and pitch controls altitude? Truth is that pilots have been arguing about this since Orville and Wilbur debated the question over a couple of beers at Kitty Hawk. AVweb's John Deakin (who claims to have been there at the time) weighs in on this issue by offering some real-world scenarios and taking a look at how modern autopilots work.
March 31, 2002
|About the Author ...
John Deakin is a 35,000-hour pilot who worked his way up the aviation food chain
via charter, corporate, and cargo flying; spent five years in Southeast Asia
with Air America; 33 years with Japan Airlines, mostly as a 747 captain; and
now flies the Gulfstream IV for a West Coast operator.
He also flies his own
V35 Bonanza (N1BE) and is very active in the warbird and vintage aircraft
scene, flying the C-46, M-404, DC-3, F8F Bearcat, Constellation, B-29, and
others. He is also a National Designated Pilot Examiner (NDPER), able to give
type ratings and check rides on 43 different aircraft types.
long promised myself I'd never get in this argument again. I have resisted
manfully (is that sexist?) Every time I start to say something on this
subject, I cover my ears and start the usual mantra, "Pink elephants,
" trying to avoid even thinking about it.
It's a little like the time we took off from Los Angeles for Brazil, and
got an unusual routing due to a major storm over Central America. As soon as
we checked in with Los Angeles Center, the clearance came, "Japanair 64,
cleared direct to El Paso, flight plan route."
Very nice, I always enjoy those long, direct routes. They really don't save
much time or fuel, but they sure do feel good. We punched it into the main
computer (the old, archaic PMS, for those who like such details), verified the
latitude and longitude, and away we went on autopilot, controlled by our
triple INS system.
was too early in the flight to sleep (I used to try very hard to stay awake at
least to top of climb) or eat (company policy was that the passengers came
first), so we sat back, relaxed, chatted a bit, and the copilot whistled a few
bars of that old Marty Robbins tune. You know the one, I'm NOT going to even
mention the name of it (pink elephants, pink elephants), but it's the one
about the (pink elephants, pink elephants) Mexican maiden and pink elephants.
I immediately howled in protest, covered my ears, and started the mantra,
but it was too late. That damned catchy little tune was, once again, imprinted
in my mind, and it took me weeks to get it out. Pretty sure I can remember
hearing it first (pink elephants, pink elephants) from the airport restaurant
jukebox in Sarasota, Florida, in 1956, where I hung out a lot mooning over the
lovely Donna Miller, the daughter of the restaurant owner, Annie. I was about
16 and Donna was about 14 and I was in love. Never even got up to bat, much
less to first base, but that is another story. I wonder where she is now
same time frame is also about the time I first heard the ancient airport
argument, probably started by Orville and Wilbur, over whether pitch
(elevator) or power (throttle) controls speed or altitude.
My efforts to avoid the subject have mostly worked, but no less than three
readers, almost simultaneously, have asked me for my thoughts on this. Since
I've only GOT three readers, I suspect collusion, probably incited by someone
who knows my antipathy for this subject.
Alas, like the song, the imprint is now there. Perhaps I can help clear it
by writing about it. This may also serve the purpose of preventing me from
criticizing the FAA, thereby keeping me more or less out of trouble for
You've all heard the arguments, always good for getting the local pilot's
lounge into an uproar. Depending on my mood, I usually giggle when it comes up
yet again, but then I get a little depressed at some of the junk I hear.
From hearing the CFIs talk about it, you'd think every time a crop duster
pulls up for his reversal at the end of the pass, he's thinking, "Okay,
remember now, pitch controls power, and altitude controls airspeed."
Well, something like that. I can never remember which is which, myself.
All seriousness aside, let's look at a few scenarios.
Take the fellow in the trainer, at the end of the runway. Runup done,
cleared for takeoff, he's sitting there pumping the yoke in and out, working
up a sweat. The tower calls, and says, "Say, we notice your elevator is
flapping, is there a problem?"
"No problem," (breathing heavily), "my instructor says that
the elevator controls speed, and I'm trying to get some, so I can add throttle
and get some altitude."
Folks, it is my firm belief that thousands of student pilots and CFIs (same
thing) have been unnecessarily confused by this worthless, STUPID argument,
and all the verbiage behind it. Even the FAA
(Ooops, I promised I wouldn't
say anything bad about the FAA. Nevermind, but check out some of the knowledge
fact is, these "rules" are not rules at all, and they are more often
wrong than right. Yes, they are right sometimes, but if you buy off on them,
then you have to memorize all the exceptions. Why not throw the whole stupid
argument out, and look at things rationally, intuitively, sensibly?
Assume you're in stable, level flight, constant speed. Your instructor says
to increase your speed by 10 knots, while maintaining altitude. Even a moron
isn't going to have to think twice about this, he'll add power, there just
ain't any other way to get there from here. As the airplane picks up speed, a
by-product will be a change in trim, and a small correction with the trim tab
would be good. I don't see how anyone can stretch that into the pitch
controlling the speed, that's getting the elephant
, er, the cart before the
horse, when it should be the other way around.
What if you're in a power-off glide, and the CFI says to increase speed?
Simple enough, you pitch down a bit to increase your speed, just as driving
down a steeper hill will do the same in your car.
Look at a boundary condition. Your evil instructor is having you practice
slow flight. The FAA today defines "slow flight," or "flight at
minimum controllable airspeed" as flight at 120% of the stall in some
places, or 10 knots above stall in others. That's kinda wussy, we used to have
to get right down to the stall, and maneuver there, but never mind, we've got
to be safe, and politically correct, and that's fine for check rides, that PTS
keeps the Inspectors and Examiners from getting carried away.
But if you're lucky enough to find a CFI that isn't scared to death of any
speed less than that, he may have you slow to a real minimum speed, and at
some point, you'll note that you have to use more power to maintain that
speed, while also maintaining altitude. This is the famous "back side of
the power curve," where it takes more power to go slower. Some confuse
this further, and call it the "area of reverse command," and the
result is often a confused student who comes away from that very scary lesson
thinking, "Gosh, if I add more power, I'll go slower."
No, that isn't the lesson. It's simply a demonstration that as you fly
slower and slower, the airplane is plowing through the air just like a
speedboat that is going too slowly will plow through the water, using lots of
power. Like the speedboat, if you add some power and get it "planing,"
you'll need less power to maintain speed (and altitude).
(And no, dear reader, this has nothing to do with the equally stupid OWT
(Old Wives' Tale) about "The Step" at cruise power. There is no such
thing, but that's another story, too.)
If you're really gentle, and the air is really smooth, you can slow the
airplane down enough that it will require full power to maintain the speed and
altitude, but this is a very delicate balancing act indeed. How do you get out
of this? You must pitch down, give up a little altitude, pick up a little
airspeed, and fly it out. Some STOL aircraft use this technique, making a very
steep, very slow, high-power approach for that last few moments, perhaps even
adding an extra shot of power to arrest most of the descent rate at touchdown.
Done properly, this is a COMBINATION of pitch and power to produce a speed and
descent rate (and angle), and that is key to this whole discussion. You must
use a combination of aircraft controls to produce the desired result!
of airplanes now have autopilots that can make the ILS approach look
disgustingly easy, and autothrottles to control the thrust. Guess what the
autothrottle computers react to? Why, errors in SPEED, of course. If the speed
drops a bit below the selected speed, the autothrottles move forward, and if
too fast, the autothrottles move back. Pretty simple, to me. As an entirely
separate matter, if the airplane goes a hair above the glide slope, the first
action by the autopilot is to pitch down to return to that path. If that
changes the speed, fine, then the autothrottles correct for that as a
secondary reaction, but that's often too quick to observe. That all seems to
be proper design to me, and it mirrors EXACTLY what a human pilot would do,
What if you're climbing? Presumably you have some specific power set, so
you can't (or don't want to) change that, right? Well, since no power
adjustment is available, the only control left is pitch, and sure enough,
you'll adjust the pitch with the elevator to hold the speed you want, whatever
that may be, and take whatever altitude or climb rate that gives you. That's
not rocket science, and you don't have to memorize a couple of rules (and
exceptions) to figure that out, do you?
What happens on a go-around? With a modern autopilot, the pilot pushes the
TOGA button ("Takeoff/Go-Around"), and several things happen in
quick-time. First, the autopilot pitches the airplane to the initial go-around
attitude (usually a fixed angle for the type), and the autothrottles come up
to full ("Go Around") thrust. Then almost immediately, the autopilot
changes to a speed mode, and controls airspeed with the pitch/elevator. That's
exactly what a human pilot does in the climb (in a jet)!
often descend at idle thrust, thrust levers all the way back against the idle
stops. What's controlling speed and altitude? Well, if you want to maintain a
given speed (let's say 340 knots), you control it with the elevator (on
autopilot), and take whatever descent rate that gives you. That's a little
willy-nilly, because it's not possible to predict the winds during descent
from FL450, it's not an exact science, and you'll need to make adjustments on
the way down. Some of us like to play games and make minor adjustments on the
way down, others just bomb on down at 340 knots, and make larger corrections
as needed when they reach the bottom. Doesn't make much difference, either
way, in time or fuel, but the former does show a little finesse.
One common method of tracking this is to knock off the three trailing
digits of the altitude above the airport (for a sea-level airport, 45,000
becomes 45), and multiply that by three (getting 135) for the approximate
correct distance from the airport to start down.
(Note this gives "air miles," or the distance to fly. If you're
on a vector for an ILS landing in the opposite direction, that usually adds
about 30 air miles to the descent.)
At 20,000 feet, you should be roughly 60 miles from touchdown. Now, suppose
you pass the 60-mile point at 340 knots, and you find yourself at 22,000? That
means you're about 2,000 feet high (or six miles "too close.") What
do you do? The ham-fisted will simply pull on the speed brake lever, making a
bit of vibration (sometimes more than a bit), to get back on profile. The more
artful will pitch down a bit, and run the speed up to 360 knots. That gives a
momentary increase in rate of descent while accelerating, which helps kill the
altitude, but more importantly, as you stabilize at the new speed, the overall
drag will be higher. Drag rises proportionally as the square of the speed, so
a small increase in speed gives a larger increase in drag. That 2,000 feet of
excess altitude will melt away pretty quickly. Conversely, if you're "too
low," pitch up a bit, slow to 300 knots for the correction, and all of
this will be imperceptible to the paying passengers.
Modern computers in glass cockpits do this very nicely, in what they call
"VPATH" mode. The computer literally draws a line from "here,
at this altitude," to "there, at that altitude," and no matter how
many turns and kinks in the track there are on the way down, that
"vertical path" will be perfect, and will be flown at the proper (or
selected) speed. By default, the computer will use "optimum" speeds
to 10,000 feet, then 250 knots below that. What's the automation doing? Why,
it is controlling the vertical path with pitch, and the speed with thrust!
is another factor that confuses the dogmatic. Some airplanes will change pitch
directly, with power (or thrust) alone! If the tailfeathers are in the
propwash, the extra air blowing over them will increase the "down
lift," and the nose will pitch up from that alone. Some airplanes, like
the 747, have the "center of thrust" well below the "center of
drag," (due to the low-slung, pod-mounted engines), and an increase in
thrust pitches the nose up quite noticeably, without any extra air going over
As Karl Malden used to say on TV, "What do you do? What DO you
You simply use all available controls to accomplish your overall purpose.
That can be as simple as pitching up or down to correct a small altitude
excursion in cruise flight (pitch controls altitude), or a small reduction in
fuel flow (power) to slow to a more efficient, long-range cruise condition
(throttle controls speed). Or, it can be very complex, to comply with
"Descend to 8,000 feet, cross WUSSY at FL 250 or above, FANNY at FL 210
or below, maintain Mach 0.84 until passing FL350, then slow to 270 knots until
further advised, expect holding at HEAVN, with a further clearance in October,
squawk 1234, change to my frequency 123.45, and the altimeter is 30.12, good
day." Now that's a challenge, even for the FMS!
Oh, and the next time you hear the local airport bums arguing about which
controls which, just toss in, "Everyone knows pitch controls power, and
speed controls altitude," and walk out. That'll raise the heat a few
Or do like I do, just cover your ears, and start chanting "Pink
elephants, pink elephants."
Be careful up there!