Airline Pilot Central Forums

Airline Pilot Central Forums (https://www.airlinepilotforums.com/)
-   Technical (https://www.airlinepilotforums.com/technical/)
-   -   Drag: Windmilling vs Dead Prop (https://www.airlinepilotforums.com/technical/98539-drag-windmilling-vs-dead-prop.html)

flyingchicken 12-20-2016 05:06 PM


Originally Posted by Captain Beaker (Post 2266111)
The graph shows 3 things:

1. Engine/Gearbox is not the cause of the drag, e.g. compare the scenario at say 30 degree pitch.

2. Flat pitch and wind milling creates very significant drag, the drag coefficient increases massively as the blade moves towards 0 degrees. This is entirely consistent with the need to feather in an aircraft cable of fine pitch. It is also consistent with safety devices needed to ensure blade angle remains outside the 'beta range' while in flight.

3. It's consistent with my own observations and the flight manual, for an aircraft fitted with a fixed pitch prop of approx 18degrees found in a C-172. i.e. wind-milling or stationary does not seem to have a very noticeable effect. It's consistent with the need to set low rpm on a constant speed single.

All of this is entirely consistent with every flight manual I have come across.

The original poster asked an interesting academic question, so maybe I provide some thoughtful input on that academic discussion. I haven't proposed anything contradictory to what you expect to find in the flight manual.

Couple of thoughts, just for fun :D

How come the propeller creates so much drag in the positive beta range, even when driven by the substantial torque of a turbine in flight idle?

While a helicopter is in auto-rotation, the blades free wheel because it is fitted with a sprag clutch. If we consider the blades to be a large windmilling prop rotated 90 degrees wouldn't the resistance from the engine be advantageous? ;)

Probably best I get my coat.:)

1 - It is important to note the context of the reference material - the graph was used to illustrate the substantial drag of a windmilling prop on a failed engine. The fact that it shows the drag curves cross over in the higher pitch angles is interesting, but without knowing the other variables - RPM, airspeed, prop dimensions and propshaft torque - it is meaningless, and very misleading - a change in any of these variables will totally change the graph.
To illustrate - note the stopped prop line represents infinte torque/friction from a stopped engine. Now lets reduce the friction so that the prop barely rotates - say 1 RPM. What would the incurred drag now be? I can tell you it will not be the windmilling drag line on the graph.
This is the interaction I was talking about earlier - the engine/gearbox friction dosen't CAUSE the drag - but is one of the inputs that will most definately affect the drag by changing the windmilling RPM at varying airspeeds. If you have an engineering background we can continue this discussion on how these higher order effects stack up.

I dont think anyone has issues with point 2 - you are pretty much reiterating the consensus.

3 - Please elaborate on your testing methods and results.

And on your 'fun' thoughts
Idle beta 'creates so much drag', but guess what happens when you go deeper into beta?

You missed the whole point of autorotation - its not done to increase drag to reduce the ROD, but to build up rotar momentum so there will be enough energy built up by the time you get to the ground to flare out for a survivable landing.

No need to get your coat - but try to consider your audience before speaking and modify tone accordingly.

Captain Beaker 12-21-2016 02:57 AM


Originally Posted by flyingchicken (Post 2266522)
1 - It is important to note the context of the reference material - the graph was used to illustrate the substantial drag of a windmilling prop on a failed engine. The fact that it shows the drag curves cross over in the higher pitch angles is interesting, but without knowing the other variables - RPM, airspeed, prop dimensions and propshaft torque - it is meaningless, and very misleading - a change in any of these variables will totally change the graph.
To illustrate - note the stopped prop line represents infinte torque/friction from a stopped engine. Now lets reduce the friction so that the prop barely rotates - say 1 RPM. What would the incurred drag now be? I can tell you it will not be the windmilling drag line on the graph.
This is the interaction I was talking about earlier - the engine/gearbox friction dosen't CAUSE the drag - but is one of the inputs that will most definately affect the drag by changing the windmilling RPM at varying airspeeds. If you have an engineering background we can continue this discussion on how these higher order effects stack up.

I posted the graph from ANA, because the text was brought up. I hadn't read it for a long time but sure enough it had the info to show that drag is not created by the engine/gearbox.

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS report no 464 which was written in the 1930's answers this in much more detail. It provides graphs for different blade angles with respect to velocity etc. It was mainly concerned with solutions for dive bombers of that period... I have another paper somewhere else.

I appreciate you understand the drag is not created by the engine, and as you point out reduced rotation caused by friction will have an effect, but there is a CUT OVER point. Precisely what that Is I don't know, thought the data seems to suggests it is somewhere around 20 degrees.

I apologise if I ruffled some feathers:D

I need to make a trip with Captain Bunsen Honeydew:eek: over the next few days but would be happy to discuss the data properly and learn something.

JohnBurke 12-21-2016 05:30 PM


Originally Posted by Captain Beaker (Post 2266696)
I posted the graph from ANA, because the text was brought up. I hadn't read it for a long time but sure enough it had the info to show that drag is not created by the engine/gearbox.

The text was quoted because you introduced it; Aerodynamics for Naval Aviators was your reference, and contradicts nearly everything you've attempted to say.

The graph you posted, from ANA, does NOT suggest anything whatsoever about mechanical drag, and does NOT make the point that you believe it makes. What it does do is show that you don't understand the information.

Captain Beaker 12-21-2016 06:02 PM


Originally Posted by JohnBurke (Post 2267268)
The text was quoted because you introduced it; Aerodynamics for Naval Aviators was your reference, and contradicts nearly everything you've attempted to say.

The graph you posted, from ANA, does NOT suggest anything whatsoever about mechanical drag, and does NOT make the point that you believe it makes. What it does do is show that you don't understand the information.

My mistake, in my very first post I had in mind the NACA data, your correct it was actually myself that introduced ANA in subsequent post. Must have been all the ad hominems that caused the confusion :mad:

But what say you about the graph in any case? What is your the explanation for the two scenarios at 30 degree pitch?

FlyingChicken makes a relevant point about other variables, perhaps the graph is not representative when other factors are considered. Fair point, perhaps look for better data...

If you care you can download NACA report 464 from NASA website
and review that data.

Captain Beaker 12-21-2016 06:09 PM

Regarding 'fun'

Originally Posted by flyingchicken (Post 2266522)

You missed the whole point of autorotation - its not done to increase drag to reduce the ROD, but to build up rotar momentum so there will be enough energy built up by the time you get to the ground to flare out for a survivable landing.

Never been in a helicopter, but I would have thought BOTH the glide and flare were important... Its important in a autogyro. It was just an illustration, perhaps a poor attempt at sarcasm.


Originally Posted by flyingchicken (Post 2266522)

Idle beta 'creates so much drag', but guess what happens when you go deeper into beta?
.

Well I would expect the FCU to command additional fuel to the engine?

flyingchicken 12-21-2016 08:12 PM

Thank you for the link to the NACA paper. It is nice to see some actual data. I would be interested to see the other paper you mentioned if you manage to locate it.

My thinking is that the "cutover" pitch you mentioned will vary dependent on prop dimension, relative wind velocity, drag/torque from internal engine friction, and resulting prop RPM. Not sure if there is a way to solve this analytically.

Keep in mind this is an airline pilot forum. The way you are going about analyzing the problem is beyond the variables a pilot can control from the cockpit. Either a prop can be feathered or it is stuck. Either the engine will windmill or it is seized.

JohnBurke 12-22-2016 04:43 AM


Originally Posted by Captain Beaker (Post 2267301)
Never been in a helicopter, but I would have thought BOTH the glide and flare were important... Its important in a autogyro. It was just an illustration, perhaps a poor attempt at sarcasm.

With a couple of exceptions, there's no collective or blade angle change in an autogyro, and whereas the autogyro runs in autorotation at all times (and flares by changing rotor disc angle, not by pulling collective (altering all blade angles simultaneously), and whereas the autogryo rotor disc has no connection to a transmission or engine, you can't correlate the autogyro with the helicopter.

The helicopter collective must be lowered and the rotor clutch disengaged (manually or automatically, depending on the system), to prevent loss of RPM (which will occur if there's drag on the rotor disc through mechanical means). flare at the bottom end in training will be power and collective, and in an engine out, final flare involves use of the collective.

Rotor blade inertia is crucial in a helicopter as is RPM. Increasing blade angle by pulling collective when in an autorotative state will decrease rotor RPM. This is not an issue with the autogyro, in which use of the cyclic to vary the plane of the rotor disc is the only option, other than adding power through an independent, fixed pitch propeller. (in nearly all cases)

Captain Beaker 12-22-2016 11:20 AM


Originally Posted by flyingchicken (Post 2267359)
Thank you for the link to the NACA paper. It is nice to see some actual data. I would be interested to see the other paper you mentioned if you manage to locate it.

Will do in the next few days, will also post some of the graphs from the NACA report with annotations, for discussion...


Originally Posted by flyingchicken (Post 2267359)
My thinking is that the "cutover" pitch you mentioned will vary dependent on prop dimension, relative wind velocity, drag/torque from internal engine friction, and resulting prop RPM. Not sure if there is a way to solve this analytically.

Same of those factors are there

I'll take a quick stab, the graph in figure 4, is were all the data is combined into a working graph.

The graph gives negative thrust coefficients (i.e. Drag coefficient) for stationary blades of differing pitch. Not surprisingly the drag coefficient remains constant with variation in speed.

You can work out the drag coefficient from a free wheeling propeller i.e. zero torque at differing pitch, this too is constant. You do this 'by projecting down from the point of zero torque to the appropriate thrust curve' then moving across for the drag coefficient.

To work out the correct solution for a windmilling prop with friction, a friction/torque curve is needed, however the actual result should lie on the respective curves so many quantitative comparisons can still be made.

I will try an illustrate this later... Don't forget to consider how a CSU effects this.


Originally Posted by flyingchicken (Post 2267359)

Keep in mind this is an airline pilot forum. The way you are going about analyzing the problem is beyond the variables a pilot can control from the cockpit. Either a prop can be feathered or it is stuck. Either the engine will windmill or it is seized.

That's true but the problem can be evaluated beyond the pilot/control limitations to further understand what's going on, is this not how this thread started?

This is the data relevant to the OP question:

fixed pitch 17degrees @100mph drag in pounds

stationary at 88° (i.e. feathered) 5.8lbs
free-wheeling at 17° 60.1lbs
stationary at 17° 94.4lbs
dead engine wind-milling at 17° 101.1lbs

Consider the scenario with C172P with an 18degree prop I believe I said there would not be much difference...

Anyway don't take my word for it read the data yourself, I could be a dog, this is after all the internet.;)

Captain Beaker 12-26-2016 12:53 PM

Following up from my last post, this is figure 4 from NACA report 464 available from NASA. I have added some annotations.

https://s29.postimg.org/qeiamh85z/drag.png

To work out the negative thrust coefficients (i.e. Drag coefficient) for stationary blades of differing pitch find where the blue line intersects the relevant curve. The blue line represent a stationary blade.

To work out the drag coefficient for a free wheeling propeller start by finding where the red line intersects the relevant curve (red line is zero torque). Then 'by projecting down from the point of zero torque to the appropriate thrust curve', move over to left for the drag coefficient.

To work out the correct solution for a windmilling prop with friction, i.e. one with an engine directly attached a friction/torque curve is needed, however the actual result should lie on the respective curves so many quantitative comparisons can still be made.

Don't forget to consider how a CSU effects this. The CSU will move the blades to a finer pitch, and even at the lowest RPM setting may result in a fine pitch.

What this graph tells us:


1. Energy used to turn over the engine, or the resistance/friction of the engine is the not the cause of the drag. It does have an effect, depending on the blade angle it can make it substantially greater, or a minor difference or in the case of a very flat blade angle (> 7degrees) engine friction can actually reduce the drag. Energy used to turn the engine is a red herring.

2. Fine pitch produce very high drag coefficients, higher than a stationary prop, and far high than a feathered prop.

3. For the blade angles 12,17,22 a reduction in negative torque results in reduction in drag, i.e. less friction is better, this is consistent with flight manual check list for closing the throttle in a piston engine aircraft.

4. Blade angles < 7 degrees, an increase in negative torque results in a reduction in drag, positive torque is required for maximum drag.

Take all this with a grain of salt, but none of this contradicts any flight manual or training manual text, with regard to pilot actions. Feathering is imperative to reducing drag when propeller is capable.

JohnBurke 12-26-2016 05:32 PM

Are you even a pilot? Your posts would suggest not. You appear to be attempting to look up material on the internet and make it fit what you think is correct, but is not.


All times are GMT -8. The time now is 12:57 PM.


Website Copyright © 2026 MH Sub I, LLC dba Internet Brands