Mechanics of autorotation

[SquirrelFlight]

Hi there - I'm still trying to wrap my head - conceptually - around autorotation. Specifically what happens in the driving range of the main rotor. I've always thought of lift as being perpendicular to the chord line... which is clearly incorrect. Are there any models - simple to moderate - that explain the forward tilt of the lift vector?

[67november]

Here Squirrel, this is a must read book and it'll answer ALOT of your questions on the mechanics of flight.

 

http://www.faa.gov/library/manuals/aircraf...a-h-8083-21.pdf

[SquirrelFlight]

Here Squirrel, this is a must read book and it'll answer ALOT of your questions on the mechanics of flight.

 

http://www.faa.gov/library/manuals/aircraf...a-h-8083-21.pdf

 

Thanks for the link - had I known about that, I could have saved the $15 or so I spent on the hard copy. Unfortunately, this doesn't have nearly enough detail. I'm looking for physics of fluid flow around the airfoil that leads to autorotation. I do own a couple of books on theory, but they treat autorotation rather abstractly - I was looking for something more concrete with more details.

[67november]

principles of helicopter flight by W.J.W

 

this MAY give you a better answer, can't say for sure as I'm waiting for my copy to arrive.

 

having the hard copy of that link is a good thing, you can break out the highlighter and mark it up to your hearts content.

Edited January 9, 2008 by 67november

[apiaguy]

have you tried Ray Prouty's "Helicopter Aerodynamics".."More Helicopter Aerodynamics"...and "Even More Helicopter Aerodynamics"??

Most are available thru amazon.

[SquirrelFlight]

I have Helicopter Theory by Wayne Johnson, and.... another Helicopter Aerodyamics I wasn't able to find I'm Amazon... I don't think it's the Prouty book as mine is paperback and Amazon only shows a hardbound edition. Either way, most of them develop autorotation as part of the power vs induced flow curves. I'm looking for something that's a little bit more of a first-principles approach.

 

Thanks, though :-)

 

 

And yes, there is something nice about having a book you can keep in a bag and flop open anytime you want to read it ;-)

[SquirrelFlight]

I just checked on Amazon and I think my other book may be an older edition of Principals of Helicopter Aerodynamics by J Gordon Leishman. I'll look into the others, too. Thanks!

[PhotoFlyer]

Put simply, in the stall region only drag is produced. In the driving region lift is produced, but it is not perpendicular to the blade, it is angled forward, and pushes in the direction of rotation. In the driven region, lift is produced, but it is angled rearward, pushing against the direction of rotation. If you want some graphics I will have to scan them, and then post, so let me know.

[SquirrelFlight]

Put simply, in the stall region only drag is produced. In the driving region lift is produced, but it is not perpendicular to the blade, it is angled forward, and pushes in the direction of rotation. In the driven region, lift is produced, but it is angled rearward, pushing against the direction of rotation. If you want some graphics I will have to scan them, and then post, so let me know.

 

What I'm looking for is a first-principles description of *why* the lift vector is tilted ahead of the chord-line in the driving region. That is, for the lift vector to be tilted forward, either there must be increased surface pressure on the airfoil from below and behind, or reduced surface pressure ahead and above (or both). In normal flight, it is easy to visualize the airfoil deflecting airflow into a low pressure region above, but well behind the leading edge; it isn't so easy to visualize the low pressure region so close to the leading edge, at least not for me. Clearly this is what happens, otherwise we wouldn't be able to autorotate.

 

Perhaps it would help to mention that my undergraduate degree is in physics, and I have almost completed a master in mechanical engineering with a focus on fluid mechanics (although at the atmospheric scale, not at the aircraft scale). Hence the increased level of detail in what I'm looking for.

[Linc]

In a helicopter, lift does not act perpendicular to the chord line, it acts perpendicular to the resultant relative wind.

 

In the case of autorotation and the driving region, the upward inflow of air modifies the relative wind so that, to the blade, it appears to come from below the blade's path of flight in that region. Since the resultant relative wind is from less than directly parallel to the blade's direction of flight, the lift vector has a more forward tilt, dragging the Total Aerodynamic Force vector forward of the axis of rotation, driving the blade forward.

 

Fig 3-22 of the Rotorcraft Flying Handbook, page 3-10.

Edited January 9, 2008 by Linc

[SquirrelFlight]

In a helicopter, lift does not act perpendicular to the chord line, it acts perpendicular to the resultant relative wind.

 

In the case of autorotation and the driving region, the upward inflow of air modifies the relative wind so that, to the blade, it appears to come from below the blade's path of flight in that region. Since the resultant relative wind is from less than directly parallel to the blade's direction of flight, the lift vector has a more forward tilt, dragging the Total Aerodynamic Force vector forward of the axis of rotation, driving the blade forward.

 

Fig 3-22 of the Rotorcraft Flying Handbook, page 3-10.

 

This makes a lot of sense... of course, I'll need to ponder it a while....

 

thanks!

[Goldy]

 

one word. Wagtendonk

[blave!]

have you tried Ray Prouty's "Helicopter Aerodynamics".."More Helicopter Aerodynamics"...and "Even More Helicopter Aerodynamics"??

Most are available thru amazon.

Helicopter Seminars has the compilation of all the Prouty books as well as Shawn Coyle's Cyclic and Collective. Both are excellent.

 

www.helicopterseminars.com/books

[Eric Hunt]

Lift is only a component of the Total Reaction, which acts upward and rearward from the relative airflow - the other component of TR is drag.

 

The relative airflow is itself of two components - the Rate Of Descent Airflow, and the Rotational Airflow. The ROD airflow is pretty constant for the span of the blade, as it is attached to the machine descending at around 1300-1800 fpm. The Rotational airflow varies with the distance from the hub, being least at the hub and max at the tip.

 

So, at the hub, the relative airflow is pretty much from straight down, and at the tip it is from closer to the horizontal.

 

The Total Reaction at the hub, although leaning backwards from the relative airflow, is actually pointing forwards from the blade, so once it is out of the stalled region, it is helping to pull, or drive, the blade forward.

 

As you move out from the hub, the TR is getting bigger and more tilted back, but is still pulling the blade forward. At some distance from the hub, the TR is vertical, so is not driving the blade or dragging it back, but is providing a larger lifting force. Further out from here, the TR is pointing rearwards from the blade, so is trying to slow it down while producing lift. At the tip there is some loss of lift from the tip vortex.

 

Along the blade, then, you have the stalled region, the Driving region, and the driven region. In a steady descent, the combination of all these factors will give you a certain ROD, with a certain RRPM and forward airspeed. Change anything like the pitch setting or the airspeed and they will all change and eventually settle at a new combination.

[SquirrelFlight]

Both Linc and Eric pointed out where my error was: I was making the assumption that lift is figured relative to the chord, no to the relative wind. Having re-oriented my thinking properly, I am able to have it make sense in my mind. Thanks guys!

 

Incidentally, NASA's Glenn Research Center has some good, basic descriptions of lift, drag and all the other fluid mechanical-type factors involved in flight, along with a fairly cool page with a Java applet that allows you to enter airfoil parameters and see the streamlines and pressure curves around the airfoil.

[ben_davenport]

The aerodynamics of autos are cool. I don't have anything to add to what Linc and Eric have said. They explained it well. I do have a couple of additional variables to throw at you. How do blade twist and the shift of the stall and driving region to the retreating side affect the autorotation, if at all? I have a few thoughts, but I would like to hear what others think.

[Eric Hunt]

Tricky bits.

 

Twist appears to spread the areas further out the blade, by decreasing the AoA as you move out.

 

Forward speed shifts the whole shebang to the retreating side. Visualize looking down on the disk from above. In the hover, there will be concentric doughnuts for the regions of stalled, driving, neutral, and driven.

 

Add forward speed, and the centre of the doughnut moves to the retreating side, pushing some of the lift-producing (but dragging) section off the end. The retreating blade now has (from the tip) :

a. A small section of lifting-but-dragging

b. The small neutral zone

c. The lifting-and-driving sector, which merges into

d. the stalled sector

e. the "zero" sector of no airflow, and then

f. negative airflow, causing loss of lift and lots of drag.

 

The advancing blade has (from the centre out):

a. stalled area

b. lifting and driving

c. neutral

d, lots and lots of lifting-but-dragging.

 

The end result is an increase in lift and an increase in overall drag with increasing airspeed. Like your normal lift/drag/airspeed curves, there is an optimum speed for auto which gives an acceptable rate of descent for forward speed and RRPM. Go faster than that, you get more drag than lift, which results in a greater rate of descent and a reduction in RRPM. Go too fast, you come down too fast and the blades are approaching retreating blade stall and dangerously low RRPM.

[ben_davenport]

Thanks for the quick answer Eric. I didn't ask my question very well. What I was going for was does the blade twist increase autorotative forces as opposed to a blade with no twist and does the shift to the retreating side increase autorotative forces as opposed to a vertical auto and why?

[Eric Hunt]

Depends what you mean by "increase auto force".

 

Twist makes each area spread out further on the blade, leaving a bit less of the lifting/dragging section outboard, compared to an untwisted blade. But nobody has an untwisted blade in real life, so i dunno what it would look like in that case - very difficult to maintain RRPM, I expect.

 

Forward flight increases forces on the advancing side but decreases them on the retreating side, flapping equalizes it out. End result, more drag, harder to keep RRPM, bigger ROD.

[ben_davenport]

Here is what I was thinking. Since the blade twist speads the lift out more evenly over the blade it moves more lift into the driving region and reduces the lift in the driven region, both of which will increase the efficiency of an autorotation. As far as the shift to the retreating side, the driven region is further out on the blade on the retreating side, which increases the torque. Plus, the blade flapping down on the retreating side will create more a steeper RRW, pitching the lift further forward.

 

On a side note, I think (though not sure) that Bell 47s don't have any blade twist.

[Eric Hunt]

"Since the blade twist speads the lift out more evenly over the blade it moves more lift into the driving region and reduces the lift in the driven region, both of which will increase the efficiency of an autorotation."

 

This is only compared to an untwisted blade - and I doubt there are any in existence - even in the 1940's design B47, which has -10 degrees from hub to tip, same as a Huey. The real reason for twist isn't to even out the lift over the blade, it is to avoid the horrendous force that would be generated at the tips if there was no twist. Blades would bend up and fail due to the increased Vsquared if the angle of attack wasn't reduced by twisting.

 

"As far as the shift to the retreating side, the driven region is further out on the blade on the retreating side, which increases the torque. Plus, the blade flapping down on the retreating side will create more a steeper RRW, pitching the lift further forward."

 

Sure, the driven region is further out, but its vector is pointing backwards, adding to drag. The DRIVING region is still inside this bit, so any increased moment is overtaken by the drag on the driven region further outboard.

 

Anyway, this is all a moot point, as all blades have twist.

[JPHarrison]

I operate an Air & Space 18A gyroplane (type certiicate 1H17, issued 1965) which flies only in autorotation. Its blades have no twist, neither is any needed.

 

It is easier to analyze autorotation from the standpoint of inclination of the total aerodynamic reaction, of which lift is just a component. Obtain a table of lift/drag coefficients for the range of operational angles, apply a little trigonomentry, and the mechanics of autorotation will make themselves clear. It is confusing to refer to the driving force as a horizontal component of lift, since it is not lift any more. It is more accurate to refer to the driving force as the forward directed horizontal component of the total aerodynamic reaction, just as drag is the rearward directed horizontal component of the total aerodynamic reaction.

 

 

www.gyroplane.aero