# About how much drag does a non-operating engine create?

I think the drag of the engine is high, for most type of engine, e.g turbo fan, jet, ramjet... I guess the engine's drag is worse than a flat plate of the same area perpendicular to air stream because the engine slows the air and compress it, and also the engine has many components inside it that will interact with air.

A flat plate is very bad aerodynamically and so the engine. Even small landing wheel should be folded inside body for better drag. So an aircraft with engines turned off will have very high drag, but I can not find any estimation on this matter.

For example, an aircraft that has L/D about 15, how about its drag if engine is removed, e.g. for better glide ratio when engine failure?

Can you estimate this drag?

• I don't know much about aerodynamics, but see if you agree with this simplification. The non-operating engine lets some air through while the flat plate lets no air through, therefore in general the flat plate creates more drag. I think it is not important what happens to the air inside the engine. – James Dec 2 '15 at 12:30

An inoperative engine creates much less drag than a flat plate of the same cross section. According to Sighard Hoerner's Fluid Dynamic Drag, the drag coefficient of a flat plate is 1.17. An engine nacelle has rounded intake lips which help the flow to stay attached while flowing around the nacelle. The closest of the generic bodies in the table below would be the sphere (drag coefficient of 0.47).

Figure 33 from Sighard Hoerner's Fluid Dynamic Drag, Chapter 3. Left column: Bodies of rotation; right column: Cross sections of 2D-bodies.

Much depends on the detail of flow separation at the forward corner, and here modern engines are rather good. If the flow stays attached, drag will be much lower than with the massive separation around and behind the flat plate. Air flowing out from the inside and over the corner of the flat plate will need some space to "turn around", effectively increasing the blocked cross section that the outside flow experiences.

Note that the reference area for all values here is the cross section of the body perpendicular to the flow, while the reference area for airplanes is their wing area.

Depending on the type, the diameter of one GE90 engine is 134 or 135 inches. Two of those power a Boeing 777 which has a wing area of 427.8 m². The ratio of engine frontal area to wing area is thus 1/23.335. The (estimated) drag coefficient of the nacelle of 0.5 would shrink to 0.0214 when referenced to the wing area. This would increase the total drag by about 50%, so instead of an L/D of 18 without engines, the airplane would have an L/D of 12 with inoperative engines.

• I know that the case of engine is smooth, but IMO you don't analyse the stream that run through the core. If the engine is turned off, that stream will rotate the blades of the turbine, and while the core is rotated, it will compress air, which may consume much more energy than a flat plate – user2174870 Dec 1 '15 at 18:24
• @user2174870: no, the core will rotate slowly and with separated flow on the blades. No significant compression will happen, and the flow rate is insignificant. Only the fan might spin a little faster, but again without compressing anything and much less flow rate than in operation. As a first approximation, you may assume that the engine blocks the air from passing through it. With propellers, this is slightly different, because a non-feathered propeller will crank the engine, which will indeed compress the air in the cylinders. A spinning propeller will cause more drag than a stopped one. – Peter Kämpf Dec 1 '15 at 21:37
• When an imperative engine is being shuttled by an aircraft do they plug the intake or lock down the fan or anything? – TomMcW Dec 2 '15 at 2:29
• @TomMcW: They are just hung under the wing, and as far as I know they do not need to be locked down. See here for the SE question on the topic (which does not go down to those details, however). – Peter Kämpf Dec 2 '15 at 7:25

There has been some studies carried out for calculating the windmilling drag (of turbojet) engines and these indicate that the windmilling drag is in the order of 10% of the net thrust. In modern high-bypass turbofan engines, the figure will probably be higher.

NASA simulations estimate the baseline drag coefficient of engine for a medium sized aircraft as 0.31, with additional drag based on Mach number.

Drag due to windmilling, Image from Validation of an Integrated Airframe and Turbofan Engine Simulation for Evaluation of Propulsion Control Modes by Jonathan Litt et. al.

There are a number of studies that deal with the windmilling characteristics of turbojet engines. In general, the effect of removal engine drag in the baseline L/D will not be very high in all probability.

You're also correct that the drag is more in case of windmilling compared to locked rotor. The drag in that case is in the order of 40% of the windmilling engine.

However, there are multiple problems with the premise that the engines would've to be dropped in case of engine failure.

• Engine failure is pretty rare for a system like engine jettison to be incorporated. In addition to the weight increase, there is the problem uncommanded operation.

• Aircrafts are certified for OEI condition. Nobody is going to drop a costly engine (note that multiple engine failures are extremely rare) because it stopped due to a minor problem. Better to fix it once the aircraft touches down.

• Even after all this , where are you going to drop the engine? unless it is over the ocean, there is no certainty that the falling engine will not cause any damage.

• Assuming you drop the engine and it doesn't damage the aircraft, you've to trim the aircraft to adjust for the new c.g. Also, aircraft will be unbalanced if you drop one of the engines, requiring more control force. Also, the wing twist will increase and the wing will bend up.

• What is a drag coefficient worth without the reference area? This answer is incomplete. – Peter Kämpf Dec 1 '15 at 16:58
• Also, the engine helps to relieve the bending moment of the wing, dropping it will increase the loading on the wing. – ROIMaison Dec 2 '15 at 10:22
• @aeroalias, you mention the value of 0.31, but I don't see that value in the graph? Is it mistake, or is it referring to another case? Some clarification would be helpful :) – ROIMaison Dec 2 '15 at 10:28
• @ROIMaison 0.31 is the baseline value. The graph shows the additional value due to Mach number. Note that the graph begins at Mach 0.2, which is about the takeoff speed for the (simulated) aircraft. So, for eg, at Mach 0.4, the value will be 0.31+0.185 = 0.495 – aeroalias Dec 2 '15 at 10:31
• Ah yes, now I understand. Thanks – ROIMaison Dec 2 '15 at 10:32