There are numerous questions and answers about the ability of an APU to provide enough thrust to maintain flight. In contrast, my question is whether the APU, by itself, provides enough thrust to move the plane on the ground. This is what motivates this question --

Back in the 1980s I often flew between Bangkok Thailand and Chiang Mai. It was normally a non-stop. But on one particular flight, it made a stop at a tiny intermediate airport. Looking out out the window (passenger window) I saw that the vehicle parking lot abutted the tarmac with no intervening separator. The parking lot was graveled, and considerable gravel had migrated to the area around the plane.

This seemed to me to be a serious safety hazard for ingestion into the engines, and I wondered whether the APU, by itself, could produce enough thrust to move the plane clear of the gravel.


  • $\begingroup$ Yes, I meant "move." Also, the airport was too small to have a tug. $\endgroup$ Commented Mar 18, 2021 at 16:42
  • $\begingroup$ With absolute certainty, no. The "thrust" is too little by orders of magnitude. Enough wind could get the airplane moving if it wasn't chocked and the brakes were released, but not the APU. $\endgroup$
    – Ralph J
    Commented Mar 19, 2021 at 0:03
  • $\begingroup$ An APU is a gas turbine but not a propulsion engine, it's a power generator (like this one). It produces no thrust at all, all power is used to turn the shaft and is converted into another form of energy, electricity or fluid pressure. $\endgroup$
    – mins
    Commented Apr 1, 2021 at 18:08

2 Answers 2


I can't reference to a document, but I can say with reasonable certainty that the answer is no, because like any turboprop or turboshaft, the APU has nearly all of the energy of its mass flow extracted for bleed air, or for torque production purposes for electricity, so that there is little thrust energy available. Turboprops typically get maybe 10% of the total energy available from the engine as exhaust thrust.

So using that number for an APU in an airplane the size of a 737 that is making, say, 650 shaft horsepower (just an educated guess; the actual number is surprisingly hard to find), it's producing a thrust equivalent to roughly 65 hp statically out its tailpipe, which equates to perhaps 2-300 lbs of static thrust (I recall that it's something like 4lbs thrust = 1 HP in the static case) at most. That would be if the tailpipe was even optimized for making thrust in the first place (tapered to accelerate the flow), which they are generally not.

So, figure on a few hundred pounds of push at most, probably quite a bit less, which isn't going to move a 737 unless it's on some kind of near-frictionless surface, or in outer space.


The APU (B737 or similar) is designed to provide air for engine start, air conditioning, etc. as well as run an electrical generator for on board systems. The power/energy created by an APU is not designed to be discharged for thrust, but for use in operating internal systems of the aircraft until a main engine is started and the APU is no longer required.

If an APU did generate meaningful thrust, the creation of that thrust would take away from its (the APU's) efficiency in running on board systems (pneumatic, electrical). Also, after starting the APU (at the gate, for example), chocks removed and during push-back, I've never experienced any resistance indicating thrust being generated at all. Based on my experience, it does not generate enough thrust to move the plane on the ground (as your question refers).

Also, the APU is on or off and the crew does not manage (with a thrust lever or similar) the power being generated.

Perhaps some day in the future, APUs or similar use auxiliary engines may be designed for dual use (provide some thrust for taxi and also run on board systems) in order to provide fuel savings.

  • 1
    $\begingroup$ It's much more efficient to add an electric motor to the gear and drive it from the APU generator than make use of the APU thrust. Jet engines are horribly inefficient for thrust at low speed (their efficiency increases with speed). $\endgroup$
    – Jan Hudec
    Commented Mar 19, 2021 at 12:13

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