[QUOTE=tdracer;10604835]I've been thinking about this, and I think the answer is that when a fan is windmilling, the fan blade airfoils are basically working as intended - the angle of attack of the blades is within the range that the flow won't separate and the airfoil stall. Hence the pressure loss through a windmilling fan is small. If the fan is locked, the incoming air is hitting the fan blade at a very high angle of attack - beyond the stall angle - so the fan blades are effectively stalled with the associated large pressure drop (i.e. drag). So there is a greater pressure loss through a locked fan than a windmilling fan, and that equates to more aerodynamic drag. I'm unaware of any engines that drive accessories off of the LP rotor so that doesn't come into play (the core would be a different question). {/quote]

Not quite. In both cases, locked and windmill, the angle of attack is on the convex side rendering the blade stalled.The differences in drag are due to what is going on in the compressor which is working to pressurize the burner sufficient for re-start, if you're going fast enough.Kinda like a ramjet

No. There is so much wrong with that paragraph that I am led to the conclusion you have virtually no understanding of rotary wing principles of flight. Debunking the misconceptions contained is worth a thread in itself so I will just debunk the first line: as soon as there is a rate of descent airflow, even with the blade at zero pitch, combined with with the rotational airflow from any reasonable Nr, the relative airflow will result in the blade producing lift (aka rotor thrust in rotary parlance). The total rotor thrust of the disc will be orders of magnitude higher than any 'drag' from non-rotating blades subject to the rate of descent airflow only. There is absolutely no way that a helo will descend slower with the blades stationary, neither initially nor in a steady state auto.

Seen from an energy point of view...

To get the engine spinning, you need energy.... and that energy comes from "stored energy" (in case of a dead engine), that stored energy will be the altitude below the aircraft, or the fuel in the tanks etc.

So no matter how you twist and turn it, if you have an engine windmilling, you're using energy to get that engine windmilling, that requires extra thrust on the other engine(s), hence extra fuel consumption.

I have to think a little more about the drag thing though, but I'm confident oggers has it nailed here.

Does this help?

https://cimg4.ibsrv.net/gimg/pprune....949b1d46bc.jpg

Another way to look at it is the potential drag energy 1m in front of either a stationary or windmilling jet engine is the same. You could say that if some of that potential drag energy is then used to windmill an engine then a windmilling engine has less drag than a stationary one, however, turbine blades are designed to push air backwards rather than be pushed by the incoming airflow so their may actually be no real benefit at all, in fact it could even be worse.

A lot would also depend on the engine size, design, location, airspeed, etc, an interesting question nonetheless.

So what you are saying, since the relative wind in relation to a windmilling engine, is that the AoA is actually "inverse", and the lift generated by the windmilling engine, is actually opposite direction of the lift generated by a normal operating engine? So the lift goes towards the rear, hence increasing the drag??

Thinking a bit about it, I can buy that, looking at the example of the gyro copter.

In a CM56 jet engine if the core of the engine seizes completely for whatever reason, will the N1 fan still rotate freely?
In the 737-300 simulator, actuation of "turbine seizure" selector on the instructor panel causes significant noise and heavy vibration and eventually the N2 indicates zero but the N1 still indicates rotation; albeit around 10 N1 only. Does turbine seizure mean engine core seizure? Or are they both different parts of the engine?

It sounds like imprecise language is used in those simulator controls. The N1 (fan) shaft and the N2 (core) shaft both have turbine stages at the back end. The rotation of the two shafts is not mechanically connected. The core being seized or simply not turning due to accessory drag at low airspeed will not directly mechanically affect rotation of the fan shaft. It sounds like the "turbine seizure" selector you describe simulates a stopped core shaft and tries to simulate some sort of N1 system mechanical failure. It's not clear what actual mechanical failure the designers were trying to simulate. The heavy airframe vibration they simulate with N2 at zero obviously is intended to represent damage to the N1 system, like out of balance or bearing failure. Failure of the N2 turbine that releases parts in the gas path often will cause heavy damage to the downstream N1 stages.

Engine fire as a result of IFSD??
 

You're right and I was painting in broad strokes. I meant "interaction" beyond that which a brick would have in its fall.

You too are right, and I shot my gun too fast. Energy goes into turning the engine, and energy goes into accelerating air - into eddies, vortices, etc., and into adding a bit of overall movement in the same direction as the craft. (Decelerating, in a reference frame fixed to the aircraft)

I'm not following. Drag coefficient (i.e., shape) is important but much more goes into it than that: namely the opposing force. By your argument, a beach ball full of air will reach the same terminal velocity as a beach ball full of lead.

In reality they both have the same drag coefficient, but the latter will have much more weight, which will oppose drag and accelerate downward faster, reaching a speed (terminal velocity) where the much higher weight is balanced by equally much higher drag.

I still maintain that an autorotating helicopter is a good analogy.
​​​​​​​
The relative wind during windmill is on the convex side which renders the AOA negative, but not necessarily stalled. Think cambered-wing airplane doing inverted flight.
​​​​​​​
Thank you for also being bothered by this! I flew a bit in a Citation where the engine synch selector positions (for which set of spools you want to synch) were labelled "Fan" vs. "turbine." I wanted to murder someone over that.
​​​​​​​

I was able to dig out the drag analysis from several years ago supporting a three engine ferry performance data approval which showed that, for a nacelle mounted high bypass engine, the nacelle with a non-rotating fan has appreciably higher drag than that same nacelle with a windmilling engine.

That's really interesting, would you care to share a bit more detail?

Unfortunately, because I don't personally own the data, I have to be careful not to release information that might be considered proprietary.

However (and this is now just me talking about the physics as I understand it), I hope most of us can agree that a fully plugged nacelle (entire fan duct and core path blocked) would have more drag than a nacelle with a windmilling fan due to the effects of 100% spillage. The nacelle with a stopped fan behaves a lot like a nacelle with a fully plugged fan because the flow through the fan duct is greatly reduced versus a windmilling engine. The stream of air that actually flows through the nacelle is greatly reduced in diameter when the fan is stopped, unlike when a propeller is stopped, due to the high solidity of the fan and the exit guide vanes relative to the flow direction and the total effect of the tortuous path the air has to follow to get around the fan blades and the exit guide vanes (which are shaped and angled all wrong for this flow case). This "spilled" air, which has to get out of the way of the nacelle rather than pass through it, causes an appreciable amount of what I believe is referred to as interference drag, but I am not an aerodynamics engineer so that may be the wrong term.

In the B-36, the jet engines are only used for the high speed dash over the target area. During cruise on piston power, nacelle doors cover the unused jet intakes. This design seems to run counter to the assumption that a blocked intake is draggier than windmilling engines.
https://images.dailykos.com/images/1...10f96f9b_b.jpg

That shape is very different from the shape of an 8 foot diameter high bypass engine with a blockage 4 feet back in the inlet. I know nothing about that airplane except for admiring it's extremely complex cockpit in a panoramic viewer you can find on the internet. Try to find the jet engine controls if you don't already know where they are!

Dave Therhino already said it, but just to be more clear: there's at least 3 configurations under the discussion:

1. Blocked/faired intake
2. Open intake with windmilling engine
3. Open intake with seized engine

And I've listed them in order of what I now think is least to most drag. (At the beginning of the thread I thought 2 and 3 were reverwed, until the few guys referenced the ETOPS data)

Your item 1 is really two configurations that I suspect are the lowest and highest drag configurations, fully faired inlet being lowest drag, and nacelle with inlet fully blocked at the fan face being highest. The fully faired inlet is a theoretical case.

Further research on the topic of spillage drag has led me to the text Gas Turbine Performance (P.Walsh/P.Fletcher) where it mentions spillage drag can range from 10 - 20 percent of internal drag in a pure turbojet, to appoximately equal to internal drag at a bypass ratio of 5:1. This does corrolate with the disconnect between the data from the old NACA reports using straight jets against what is being put forward here.
Its a good day when one sets off to prove someone wrong only to demise by his own sword!
The text references data from ESDU 81009, 84004 and 84005 - if anyone has access and could summarize it would be greatly apperciated.

Higher drag with a seized engine.

All discussion of how the windmilling fan is extracting energy and thus must be creating more drag ignores the fact that a seized engine is essentially a giant blunt body with all kinds of intake spillage going on. That effect is way more than the windmilling. (All the windmilling needs to do is overcome the bearing drag once the engine is stabilized, which is not that big)

Am aware of an incident where significant engine damage caused one of a twin to seize. Max attainable altitude dropped from the theoretical ~25kft for a normal OEI case to below 15kft.

Great discussion. I was starting to question my intuition with the back and forth. My initial response way back at the start was from simply observing that any system (in this case a turbine engine exposed to oncoming air) is going to naturally seek the low-energy, low-drag state. Think about sticking your arm straight out the car window - it takes an input of energy from your muscles to maintain the higher drag configuration instead of letting it fall back. Likewise, given the option, an unpowered turbine that is free to rotate will - because it would requires more energy to resist the force of the oncoming air. That energy is proportional to the increased drag of the seized condition. This more of a physics/thermodynamics argument, but I think it still gets us to the right answer.

From the text referenced above - locked rotor is not necessarly less drag than windmilling - it depends on the bypass ratio. In straight jets and low bypass fans the losses from the increased flow through the core more than offsets the spillage. The crossover point was mentioned to be around 5:1 bypass - right around where the last gen turbofans live.

Concerning props, it appears there is also a crossover point :
http://www.peter2000.co.uk/aviation/misc/prop.pdf

For now, I'm going stick with the thermodynamics model, and to that point I would say that the minimum drag stable point of a low vs high bypass would show up in the speed of rotation.

Actually, 5 to 1 is around where the first generation of 'big' turbofans lived - JT9D, CF6-6/CF6-50. It's gone up considerably since then with the latest generation approaching 10.

If you carefully read what you wrote in post #59 above, your paraphrase/summary does not say that the 5:1 bypass ratio is the crossover point for a locked rotor nacelle drag being more than windmilling nacelle drag, but rather that, for some nacelle design, a 5:1 bypass ratio is the bypass ratio where the internal drag of the windmilling nacelle (due to force on the fan and force on the internal scrubbed surfaces) equals the external drag of that same windmilling nacelle. If I'm reading that correctly, that doesn't tell you anything about how a locked rotor nacelle compares to a windmilling nacelle. The locked rotor nacelle has less internal drag due to lower duct flow and a lot more external drag due to greatly increased spillage. I didn't read the actual document - only your summary from it - so I may not be understanding it correctly.

Internal drag on a windmilling pure jet would be much higher than the internal drag of a windmilling high bypass engine of the same diameter. For a given external nacelle geometry at a given spillage flow and given flight condition, the external drag is constant and has nothing directly to do with bypass ratio. Internal drag varies with significantly with bypass ratio. The higher the bypass ratio of a windmilling engine of a given fan diameter, the lower the internal drag because more of the air only has to go through a fan and EGVs and not through a whole engine core. With a duct blocked at the fan face, bypass ratio doesn't matter at all because all the air is being spilled. With a locked fan passing far less air through the duct than a windmilling engine, bypass ratio has less of an effect than it does for a windmilling engine. So expressing a windmilling nacelle drag versus locked rotor nacelle drag crossover point purely in terms of bypass ratio doesn't appear to me to make sense. I suspect absolute fan and nacelle diameter also has a significant effect due to the fan area being proportional to the square of diameter and the nacelle outer circumference being directly proportional to the diameter, with larger diameter engines having a greater overall drag difference between the windmilling and locked rotor configurations.

My recollection from an airplane program involving a very large twin with 6:1 bypass ratio engines was that the nacelles were analytically shown to have quite a bit higher overall drag with a locked fan rotor versus a windmilling rotor.

Going Down.
 

Well, there's not much else we can do is there? Might as well make a movie; it could relieve the tension while waiting for the final lurch.......

Interesting thread. It seems that conditional drag is dependent on engine design, more specifically, bypass ratio.

Makes the locked rotor/windmilling issue mute. Torque imposed by the rotor locking up tears the engine off.

https://cimg4.ibsrv.net/gimg/pprune....e9064c0fbe.jpg

I've often wondered what would have happened if it was the No. 2 engine that locked-up.

Best not to get into torque and locked rotors and just stick to aero loads in this thread. Much of what has just been posted is wrong in analogy. But hey, it's a free for all so just take it where you want :)

It wouldn't have been much different. Remember where that center engine is - in the very back on the fuselage centerline - it's not in the tail. The 727 engine mounts had 'shear points', designed to kick the engine away from the fuselage in the event of a structural overload.
Besides, remember what was believed to have caused the engine seizure - blue ice ingestion. It's pretty hard to envision a scenario where a big chunk of blue ice ends up going through the center inlet...