Many light general aviation aircraft-- and perhaps many other aircraft as well-- are designed with some of amount of downthrust in the engine mounting, relative to the centerline of the fuselage. Presumably the wing incidence angle is chosen to streamline the fuselage in some particular flight regime. So why is any downthrust designed into the engine mount? Exactly what is the designer trying to optimize, by including downthrust in the design? In what undesirable way would the aircraft's flight characteristics be different if the downthrust were not there?

For example, is the intention that adding power (without re-trimming the elevator) should not make the aircraft climb? Or that adding power should not cause a change in trim airspeed? Or that adding power should not cause a change in trim angle-of-attack? Would there be a "wrong-way" change in trim airspeed when power was added, if the downthrust were not present? If so, why?

Note that one reason for some reduction in trim airspeed when power is added to climb but angle-of-attack is held constant, is given in this related answer. (See the tables-- and note that a similar reduction in trim airspeed would also occur when power was reduced to descend.) The present question anticipates that this effect may be minor enough not to be the main reason for including downthrust in the design of any particular airplane.

This question is aimed primarily at aircraft with nose-mounted propellers and conventional tailed configuration, so please be sure to address this configuration, but feel free to discuss other configurations as well.

  • $\begingroup$ This is a real question not a "trap" question; I have some idea of what the answer might be but wouldn't be able to offer a definitive answer w/o some further research! $\endgroup$ Commented Nov 14, 2020 at 14:02
  • 1
    $\begingroup$ I think you'll find that if you look at the thrust axis in flight at normal cruise AOA, as opposed to the thrust axis relative to the longitudinal axis of the plane itself, you'll find that it's more or less horizontal in level cruisiing flight. If wing incidence is zero, and the engine is canted down 3 deg, and the wing operates at 3 deg AOA in cruise flight, there's your horizontal thrust axis with no significant UP or DOWN component to it. $\endgroup$
    – John K
    Commented Nov 14, 2020 at 14:38
  • $\begingroup$ @JohnK -- thanks for the note, but would the wing incidence really typically be zero? This does not seem optimized for streamlining the fuselage at cruise speed. Why would the wing incidence not be set to streamline the fuselage at cruise speed? So I understand your contention is that fuselage is typically not level at cruise and that's why the downthrust is there, to make the thrust line parallel to the flight path at cruise? $\endgroup$ Commented Nov 14, 2020 at 23:50
  • $\begingroup$ @JohnK-- my belief is that at least in the case of high-wing single-engine airplanes, the downthrust is there to counteract a tendency for the wing's downwash to strike the tail and pitch the nose up to a higher angle-of-attack, especially at higher power settings, since the downwash is accentuated by the propwash. But I'd like to hear it from someone who has better knowledge of this either from actually working in a/c design, or from more extensive research than I've done at present. $\endgroup$ Commented Nov 16, 2020 at 13:11

2 Answers 2


With regards to John's comment, we should remember the other player applying torque to the Center of Gravity, the horizontal stabilizer/ elevator/ stabilator/ trim tab. Ideally both the thrust line and the horizontal stabilizer will be 0 degrees to the horizon with the wing at optimal cruise AOA.

Under power, relaxing static stability with a little competing downthrust will lessen the need for excessive trimming any time thrust is changed, and also leaves static stability full on when the plane powers down to land.

This is a very safe and widely used application of downthrust for GA aircraft. Because the cosine of the small downthrust angle is very small (cos 3 degrees = 0.9986), effect on the horizontal thrust component is minimal.


Most light aircraft sit slightly nose-up in flight. This simplifies the design parameters for the wing mounting, as the wing needs to angle slightly up. The designer seeks to maximise thrust by angling the engine down so it points straight forward.

On a high-wing type, angling the engine down also adjust the thrust line so it passes closer to the centre of pressure. This reduces trim changes with engine power.

However on a low-wing type this requires the engine to be angled upwards. But angling it up too much moves the thrust line lower still, so that in an engine-out situation the nose will drop to help maintain airspeed and avoid a stall. The angled thrust also introduces a vertical component, i.e. lift. This is at the expense of forward thrust. For a small angle, the loss of forward thrust and airspeed is insignificant, and the extra lift actually offloads the wing a little bit, allowing a reduction in drag so that the plane may actually fly a fraction faster. But for a significant angle the thrust loss increases more than the wing drag decreases and the plane slows down worse than before. So the angle should not be overdone.

  • $\begingroup$ It seems like you are suggesting that the typical situation involves upthrust, not downthrust. Is this really true? $\endgroup$ Commented Nov 14, 2020 at 23:53
  • $\begingroup$ @quietflyer I take your point. I'll add something to my answer. $\endgroup$ Commented Nov 15, 2020 at 12:14
  • $\begingroup$ @Guy Inchbald "for a significant angle the thrust loss increases more than the wing drag decreases. It gets good when the numbers are run. How about a calculation for our generic GA 10/1 L/D ratio, 5 degrees up thrust? It would be sin (5) for vertical up thrust, cos (5) -1 for horizontal thrust loss. But sin (5) wing deflection drag spoils it unless we keep lift vertical with leading edge slats. +1 for some numbers. (Slats may chip in even more lift from increased camber)! $\endgroup$ Commented Nov 15, 2020 at 18:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .