"Disc'ing" drag

[hugh flung_dung]

I was recently asked why the drag from a windmilling prop was greater than that from a stationary prop. No problem (I thought) and proceeded to give the standard vague explanation - the stude then asked for a bit more detail and I ended up falling on my sword and promising to have an answer by the next time we met.

So, are there any budding aerodynamicists out there who can give a rigorous explanation?

[skyvan]

The theory, as far as I understand it, is as much a matter of aerodynamic drag, as "discing" drag. Basically, the windmilling prop is being driven by the airflow, the force required to turn the prop has to come from somewhere, and that force is actually the extra drag, when compared to a stationary prop.
Just as an aside, when I flew a Dash7, we had to be very careful on the flare, as when the thrust levers were retarded to idle, the increase in drag could kill 10kts if you were still flying. That was 4 huge props that generated drag, enough to slow an 18 000kg aircraft very rapidly.
If the student doesnot understand, tell him to go and research it, and come back to you with the answer, otherwise, the book "Flight without Formula" by A.C.Kermode will probably have a good (clearer) explanation.

[Luftwaffle]

Consider the drag forces on the propeller. The surface area the propeller presents to the relative wind does not change because the propeller is turning, so form drag would seem to be the same. Meanwhile, interference drag would be greater for the turning propeller, because of all the turbulent air flow around it.

Or is that the 'standard vague' answer?

[pilotbear]

Also, a)the propeller when rotating is creating lift in a forward direction pushing against direction of motion of the aircraft and b) as we all know the creation of lift means induced drag so the propeller is also trying not to turn hence even more drag.
Plus the friction of the engine which it is trying to turn.

I hope this makes sense

------------------

[4dogs]

Folks,

The initial question was about the difference in drag between a windmilling prop and a feathered prop. Discing, as we commonly use the term, is a different version of the normal windmilling situation.

A feathered propeller presents a stationary profile to the breeze. It generates form drag. It does not generate lift and thus no induced drag. It cannot generate positive thrust.

A windmilling prop is not (normally) being driven by the engine. It rotates because it generates a net rotational force (lift component) in the direction of rotation (positive component). The rotational force is overcoming the machinery drag of the engine and gearbox only and, importantly, is a force at right angles to the direction of flight.

In helicopter parlance, the propeller is autorotating. That happens because the blade is twisted and the angle of attack (and therefore lift component)on each blade element varies as a function of distance from the hub (rotational velocity), forward velocity and incidence (blade twist). The alpha varies from positive to negative along the blade. The rotational force is a net positive component because there are negative components due to stalled areas of the blade (too much alpha) and non-lifting areas (not enough alpha).

However, the drag we are referring to is in the direction of flight, not the direction of blade rotation. It comes about because the propeller is generating negative thrust. The twisted blades, although rotating, produce an overall drag component in the flight path axis from the stalled sections and the non-lifting (or even negative thrust producing) sections.

The severity of the windmilling drag is related to the windmilling (rotational)speed. Quite often, a feathered prop will continue to rotate slowly, however the drag is not very high. As we unfeather the prop, we are mechanically changing the alpha, thus generating some positive rotational force but substantially increasing the drag. As the prop speeds up, we approach windmilling drag - the steady state drag we get from extracting all of that energy from the airflow to drive our propeller. In most cases, the windmilling RPM is fairly low because the propellor has mechanical stops that prevent the propeller from getting ultra-fine.

In my experience, we tend to speak of discing in terms of turbo props that have either beta ranges or reversing props. These props have stops that permit much higher windmilling speeds because you cannot afford to lose significant propeller RPM when you select idle if you are then about to select beta or reverse thrust. They generally govern the engine to maintain propeller RPM. Hence, the drag increase when you select idle with those propellers is a quantum leap over that which you experience selecting idle in a (typically piston-powered) non-reversing propeller.

Comparisons would best be made by describing discing as entering a pitch range that generates significant negative thrust while the propeller is still autorotating.

The next step is to select even finer pitch which results in lots of negative thrust and a net rotational drag which must be overcome by engine power - "reverse thrust".

Have I clarified or confused? I much prefer diagrams but the techology to do that on PPRuNe escapes me.



------------------
Stay Alive,

[email protected]

[rolling circle]

We're getting off the point here. The original question referred to a stationary prop vs a windmilling prop, no mention of feathering and 'discing' is a red herring.

I assume we are referring to the difference in drag, in something like a PA28/Cessna 150, between an non-mechanical engine failure where the prop is still rotating and an engine seizure where the prop is stationary but not, of course, feathered.

As 4dogs suggests, this is difficult to explain in simple terms without the use of diagrams so, if you check your e-mail, h_f_d you may learn something to your advantage.

[4dogs]

Rolling Circle,

Damn, I forgot Rule No 1: RTFQ!!!!

(Sorry, that's Read The Farking Question for the sheltered ones...)

Just when I had convinced myself that I was on the right track, I sense the need to also seek enlightenment. If it is not too much bother RC, could you send your Email to me as well please?


------------------
Stay Alive,

[email protected]

[Dan Winterland]

Never mind the theory. I had to turn the engine off on a Chipmunk once. I found the drag greater with the windmilling prop,(as opposed to the prop being driven at idle), and the resulting forced landing pattern had to be modified to ensure I didnt land in the undershoot. I was in regular practice at forced landings - I was an instructor on the Chippy at the time, I did many each month, so I noticed the difference.

The fairly coarse prop on the Chippy ensured that the prop did not stop until on the landing roll. I'm sure the finer examples on Cessenas/Pipers would actually stop, but I can only surmise that the drag would be similar to the Chippys', or perhaps even greater.

After a crewroom chat with my fellow instructors, we concluded that the prop driven at idle would present a smaller angle of attack and therefore less drag. (working in reverse of course.)

Why did I have to switch trhe engine off in the first place, that's a different story which I won't relate on the grounds that I will make myself look stupid!

P.S. The pattern I used was the military constant aspect procedure. Far better than the civilian one. Anyone care to comment?

[Roller Merlin]

DW
A bit off the original topic, but you did request comments on the traditional Vs constant aspect Forced Landing technique.

I utilise both on a daily basis and vouch for each within context. The standard square pattern is good for a student to learn the basic techniques because it permits the student latitude of convergence on base to fix up a shocker. The student turns onto square base and rolls out, then assesses the aimpoint with wings level. Undershooting is picked up quickly if the aimpoint is moving away (moving up the windscreen and looking flat) on base. It is simple and particularly useful when you are flying an unforgiving aeroplane with a low aspect ratio and poor glide performance. If the wind catches you out on base in these types, you are struggling to make it back to the runway. Hence the first snappy turn onto base.

The constant aspect approach you refer to is ideally suited to an aircraft with good gliding performance so you can finese the technique. A constant angle between your eye and the runway surface is flown - like a final with the aimpoint displaced to the side - and angle of bank adjusted finely to keep the aimpoint in perceived constant spiral plane (helix)to the eye, progressively tracking around as the aircraft turns onto final, terminating as a fixed glidepath aimpoint out the front. With an aimpoint being say 500ft to 1000ft in, the flap then is utilised soley to bring the aimpoint back to the keys. Alternatively, the spiral can be aimed at the threashold throughout, accepting slight IAS increases, and the flap is used as a speedbrake to achieve Vat.

With the chippy's nice wings it must be a good glider and ideal for the later training.
For experienced pilots the aspect technique is excellent if understood and practised frequently.

Cheers

[pilotbear]

If the prop is windmilling at idle then the majority of the work in rotating it will be still be done by the engine, so there is not as much drag as if the the engine was stopped.
If it is stopped it has the compression to fight against and the extra friction from the oil getting colder.



------------------

[Dan Winterland]

RM,

The constant aspect approach works well no matter how well your aircraft glides. If anything, it's probably easier with a steep glide angle as the aspect changes are more easily recognised when you are rapidly plummeting earthwards. The chippy did glide well and constant aspest PFLs were easy. The Bulldog had half the gliding performance which made it easier to PFL as it was more of a 'by numbers' exercise. the Tucano was a pig as it had too much glide performance and getting the height off was a problem. The Hawk is easy, again another numbers exercise.

As a measure of glide performance, you have to look at the hi-key heights required. (For those who don't know the procedure, hi-key is the overhead the threshold height at which you start the pattern). The Chippys was 1500', Bulldog 2500', Jet Provost 2500', Tucano 2500' (too high for that aircraft, so I tended to use 2000') Hawk 4500', and Harrier 12500'.

I wrote an article for the CAA GA safety magazine some time ago about the constant aspect approach. At the time, there were a couple of panel examiners who were keen to introduce it into the CAA sylabus. The stats spoke for themselves. The square pattern in GA had a success rate of under 10%. The military way - over 70%. Unfortunately, there was a lot of opposition, mainly due to a perceived veiw that it would increase the hours required to do a PPL. IMHO, the main reason for the success of military PFLs is that military pilots are required to stay in current practice - one every 2 months as a minimum if my memory serves me correctly. I used to practise far more, which meant that when I had do do one for real, it worked.

For those of you who haven't seen the constant aspect PFL, I implore you to go flying with an ex-military QFI and try it. You will probably like it. I ended up teaching an flying club instructor how to do them once when he was supposed to be giving me a check ride.

[Constable Clipcock]

Maybe a simple comparision is in order: ask "What makes a Ram Air Turbine function?" Answer: "A transfer of energy". The aircraft's kinetic energy is being transfered into keeping that prop and all its associated hardware rotating. It's got to come from somewhere, right? No energy transfer, no rotation.

Dan Winterland: That was an outstanding explanation on the merits of doing forced landings the military way. I saw that same method in some USAF flight training manuals I got hold of a few years back (I've never heard it called the "Constant Aspect" technique until you mentioned it, but the concept's the same). Definitely beats the hell out of the "square corners" technique the FAA's Flight Training Handbook teaches. Simple enough that I was able to figure it out from the text and then use it the next time I had the power pulled during a dual session.

The military 360º Overhead Approach seems to work pretty well for light aircraft also. For the Citabrias and Decathlons I used to fly, I personally felt more comfortable using this method than the civilian Rectangular Pattern.



------------------
Contrabanderos may run, but they can never hide!

[Checkboard]

I think the question concerns fixed pitch propellors, as such talking about feathering, of coarsening of the prop is simply not relevant.

If the fixed pitch prop is stationary, it is presenting something close to a 90° angle of attack to the airflow, and quite a bit of drag is associated with that.

As the prop rotates faster and faster, the angle of atack reduces, and so does the subsequent drag.

If the engine is producing any power at all, the prop may rotate fast enough for a 0° AOA - this is the "zero thrust" that is set for twin engine practice to simulate feathering.

Increase the power past this point, and the prop's AOA goes from negative to positive, and the prop begins to produce thrust.

The argument about the airflow providing the energy to rotate the prop is correct (where else would the energy come from), but it is incorrect to state that a stationary prop is not extracting any energy from the stream, it is just wasting this energy in form drag.

[rolling circle]

Checkboard - not exactly.

If the propeller is stationary it will produce only form drag, the amount depending upon the profile of the blade faces. If, however, the propeller is windmilling, the combination of its rotational velocity and the TAS of the aircraft to which it is attached result in a relative airflow which produces a negative angle of attack and, therefore, an aerodynamic reaction. The component of this reaction parallel with the longitudinal axis of the aircraft adds to the drag forces and might be termed 'windmilling drag' whilst the component in the plane of rotation of the propeller acts in the direction of rotation and represents the force which keeps the propeller rotating against the compression of the engine, what might be termed the 'windmilling force'. Thus a windmilling propeller produces significantly more drag than a stationary one.

I think that's the way Harvey Spirit described it but it really is easier with diagrams!

Dan Winterland - good to know you're still around, haven't seen you since Scampton.

[This message has been edited by rolling circle (edited 13 July 1999).]

[hugh flung_dung]

It's certainly true that for a constant prop pitch and fixed IAS:

- a high engine RPM gives a positive prop AoA and therefore positive thrust
- some lower RPM gives zero prop AoA and therefore zero thrust, and
- an even lower RPM gives negative prop AoA and therefore negative thrust

.....but that explanation holds whether or not the engine is running. The bit that I can't get my brain around is how to prove (I agree that it's intuitively obvious) that the overall drag at some (low) RPM is lower when the prop is being driven by the engine than when it is windmilling at the same RPM.

[This message has been edited by hugh flung_dung (edited 16 July 1999).]

[Checkboard]

The bit that I can't get my brain around is how to prove (I agree that it's intuitively obvious) that the overall drag at some (low) RPM is lower when the prop is being driven by the engine than when it is windmilling at the same RPM.

If the RPM is the same, then the drag is the same, engine running or not - the thing is that if the engine is running, the RPM will be greater than if the prop is windmilling.

Rolling Circle:
Prop spinning fast: thrust
Prop spinning slower: zero thrust/drag
Prop windmilling (slower again): drag

Starting to see the pattern? The slower the prop is rotating, the more drag it produces (assuming fixed pitch, of course). You assert that if it stops, the drag reduces :I don't think so!

[watford]

I have to agree with rolling circle on this one. The difference as r.c. has said before is that a stationary prop is subject only to form drag. A windmilling prop is subject to both form drag and 'windmilling' drag.

It is a fact of life that a windmilling prop produces more drag than a stationary (fixed pitch) prop. We proved this conclusively at ETPS years ago.

[This message has been edited by watford (edited 19 July 1999).]

[4dogs]

Checkers,

If the RPM is the same, then the drag is the same, engine running or not - the thing is that if the engine is running, the RPM will be greater than if the prop is windmilling.

I don't think so. Complex flows, twisted blades, variable alpha - the answer is not quite that simple.

Even with fixed blades, the same speed of rotation may be achieved as a combination of power applied and thrust or drag vectors in the plane of rotation. If we fix the speed of rotation, the blade element alpha is controlled only by the TAS (mostly) and the inflow angle. The result is that at low TAS the blade would stop due to high rotational drag so we have to use lots of power to maintain the fixed RPM. This is the same as the take-off case and produces postive thrust. At high TAS, the alpha is such that the prop tends to overspeed through windmilling (rotational thrust) requiring no engine power to maintain the fixed RPM yet producing significant negative thrust, or "drag" as we normally like to call it.

Thus the "drag" vector (along the aircraft flight path) varies quite considerably at the same RPM.

What I think stays the same is the net energy required at that RPM - it is absorbed from the airflow (TAS) or the engine or both. The energy required when the prop is stationary is quite low. When the prop is windmilling, buckets of energy are absorbed.

We need to be careful using the concepts of drag, particularly when we use it as it applies to normal wing sections. When the prop is stationary, it is just a collection of little wing sections aligned with the airflow and we can properly describe the result as form drag in the flightpath axis. However, when the prop is turning, the whole "drag" game gets messy. The blade sections are subject to both form and induced drag relative to their local airflow, but that is not the same axis as the aircraft flight path. When we resolve various forces from the propellor along the aircraft axis we are actually seeing vector combinations of blade section lift and drag - in the windmilling case, the greatest contributor is blade section lift that just happens to point backwards (in an aircraft sense).



------------------
Stay Alive,

[email protected]