HobbyKing.com New Products Flash Sale
Reply
Thread Tools
This thread is privately moderated by mnowell129, who may elect to delete unwanted replies.
Old Jan 12, 2007, 03:36 PM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
Quote:
Originally Posted by edi
Really? Also for a one-blade rotor vs. a two-blade one?
Actually I think, yes.
One blade will work but it takes counterbalancing.
Even vs. odd determines if two blade are simultaneously leading and lagging at the same time (even) or only one (odd).
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Sign up now
to remove ads between posts
Old Jan 13, 2007, 05:05 PM
Winging it >
leadfeather's Avatar
Joined May 2006
9,766 Posts
Rotor test

I tested four different rotors for pulsing. I did this by mounting the rotors one at a time on a stick that had a spindle at the top for the rotors. I would then walk through my "wind tunnel" basement.

I tested a 2,3 and 4 bladed Whopper/Twirl type blade (flat plate, fixed pitch) and a 2 bladed flexible feather-hinge blade from my Cyclone.

Both 2 bladed rotors had the most noticeable pulse (2 per rev) and seemed to be equal in pulse magnitude.

The 4 bladed rotor had very little pulse but I could feel it.

The 3 bladed seemed to be the smoothest, I couldn't really even feel any pulse although I'm sure it had to be there.

Dan
leadfeather is offline Find More Posts by leadfeather
Reply With Quote
Old Jan 14, 2007, 06:33 AM
Winging it >
leadfeather's Avatar
Joined May 2006
9,766 Posts
Photos from rotor test

Attached is a photo showing the rotors I tested for pulse.

From left to right:


1 Self feathering two blade from current Cyclone project

2 Two blade fixed pitch, flat blade from my first project;Vortex (Whopper-ish) design. Two of the original blades were removed for this test.

3 Four bladed fixed pitch, flat blade from Vortex,

4 Three bladed fixed pitch, flat blade from GyroStick Twin



I use the same spindle (1/16" music wire) and bearing (3/4" long brass tube) on all of my rotors so swapping and comparing is easy.

My test rig is just a carbon tube with a 1/16" wire at the top as shown on the far left.

Dan
leadfeather is offline Find More Posts by leadfeather
Last edited by leadfeather; Jan 14, 2007 at 06:39 AM.
Reply With Quote
Old Jan 19, 2007, 02:00 PM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
A blade in the middle of my back.

Some stuff about blades.
Two airfoils are drawn. A known property of airfoils is that the lift on the airfoil is such that you can represent it as a lift force at the 1/4 chord point and a constant pitching moment. The amount of lift varies with the airfoil angle of attack, but the pitching moment is constant for any given airfoil. For symmetrical (non cambered) airfoils the pitching moment (Cm) is 0.
The point is that you can pretty much pretend all the lift acts through this single point (the aerodynamic center (AC)). Lets assume that the pitching moment on a cambered airfoil is pretty small so we can ignore it.
Now because we understand that we need cyclic pitch to control our copter we are going to need to change the pitch of our airfoil.
Let's examine three possible hinge locations from which to vary our airfoils angle. At the trailing edge first we see that the lift is going to act to make the airfoil exert force on our pushrod, etc. to make the angle go higher than we are asking. This is a divergent situation. So we try the leading edge. Now the lift of the foil tries to make the airfoil go back toward 0. This tries to take out the control we are putting in. Now try at the AC. In this case, other than the small constant moment, we can move the airfoil easily and the lift force doesn't try to hinder us or run away with too much control. If our blades were symmetrical there wouldn't even be the small constant moment. (Now would be the time to ponder what symmetrical flybar paddles on a helicopter have to do with small servo loads.)
Ok so armed with this knowledge we mount our blades at the 1/4 chord point on our gyrocopter and this is all lined up with our main shaft.
But suppose that our blade CG is not on the 1/4 chord line? This is generally true of most airfoil construction, even with hardwood leading edge.
Shown is the top view of a blade traveling to the left. If the CG is aft of the 1/4 chord point we get what happens in the middle blade. The blade "leads". What is significant is that a large part of the blade is out ahead of the feathering axis. This causes this part of the blade to try to "run away" in the positive direction. The only thing preventing this is the torsional rigidity of the blade.
So you wise up and figure you have the problem licked. You just move the mounting hole to match the CG. But what this does is now hinges things such that the whole blade is in the divergent case. This is not an improvement.
Lets take the opposite case and assume you made a blade of styrofoam with a lead leading edge. The blades are nose heavy. What happens is that the blade instead of leading now lags. Now the blade fights your control input and the controls are sluggish. The tip tends to go back to zero and the only thing preventing this again is the torsional rigidity of the blade. Once again you try to solve this by moving the hole. But now the whole blade fights your control input.
With this knowledge in hand, the solution should be clear. For a stable blade with optimum control you should correct the blade CG to lie along the 1/4 chord line and the bolt hole and feathering axis should all lie line along this line.
The usual way to do this is to add tip weight in the leading edge of the blade. This not only CG corrects the blade but moves the blade CG outward, reducing the coning angle.

Now how can all those models be working without tip weight. There are a couple reasons. Obviously with small models the torsional rigidity of the blade is very high compared to its size. The "stack 4" blade construction also provides a very rigid in-plane resistance to leading. This means that the centrifugal force trying to make the blade lead is absorbed by the blade. Non cyclic rotors like the dual rotor type whose cyclic is input by moving the fuselage (thus moving the masts, in an indirect spindle tilt kind of cyclic ( a previous post)), have plenty of control authority with the tail surfaces far away. That combined with rigid in-plane blades and high torsional rigidity mean that CG correcting these blades is just overkill. However in a single rotor, with any kind of cyclic, especially with blades that are somewhat free to lead lag (to reduce hub stresses due to conservation of angular momentum as you will recall) CG correction of the blades can make a big difference.
Thats all for now.
There's one other blade topic, that being dealing with the constant nose down moment of cambered airfoils. That topic directly affects the blade aspect ratio and rotor diameter.... more later.
mick
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Old Jan 19, 2007, 11:37 PM
Registered User
Joined Feb 2005
53 Posts
I anxiously await your next installment...
M. Smith is offline Find More Posts by M. Smith
Reply With Quote
Old Jan 20, 2007, 06:21 PM
iPhly R/C with iPhone
iter's Avatar
Silicon Valley, Calif
Joined Jan 2005
1,724 Posts
Mick,

Thank you for this post. It clarified to me why my blades work even though they are not balanced. I understand now how the forces interact and the effects of scaling on my rigid heads.

Ari.
iter is online now Find More Posts by iter
Reply With Quote
Old Jan 20, 2007, 06:24 PM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
You're welcome. Small size definitely has it's advantages.
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Old Jan 20, 2007, 06:52 PM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
More blades...

A quick installment on blades. Look back at post 109.
The small nose down pitching moment (Cm) that I glossed over I would like to revisit.
The lift generated by a rotor is calculated by the formula
lift = Cl * 1/6 * rho * S * Vt * Vt.

All these symbols are :
Cl = lift coefficient of the airfoil
rho = air density
S = blade area
Vt = tip speed

What you need to take from this is that the lift varies linearly with blade area, but as the square of the tip speed. In practical terms if you make the blade area double the lift doubles but if you double the tip speed the lift goes up by a factor of 4.
So in many ways its best to have a high tip speed. You get the most lift, the lift dissymmetry is the least, coning is the lowest, the airfoil is more efficient, etc. etc.
On other factor that you just have to trust is that the larger the diameter of a rotor craft, the more efficient it is (this basic physics will always guarantee that a harrier will use more fuel to hover than a helicopter, even if they weighed the same.).
So we'd like a big disk with high tip speed.
To get the tip speed up we need efficient blades. Blade efficiency is governed by the airfoil itself, the surface finish and the aspect ratio (lenght/width) of the blades. Smooth, high aspect ratio blades are the best. So your maximum efficiency is long skinny blades and a high tip speed.
So something limits how long and skinny a blade can be. In full sized craft the tip speed approaching supersonic is one limiting factor, not a problem in models. The second limit is structural.
The longer and skinnier a blade is the lower its ability to resist twist. But you say, we've CG corrected the blades, why would they twist. The answer is that constant nose down pitching moment of a cambered airfoil. The nose down moment tries to twist the tip down, even when it has twisted to zero lift (the pitching moment of a cambered airfoil is always present, even at 0 lift).
The torsional rigidity of the blade is what stops the blade from twisting and the longer and skinnier the blade the less it can resist this twist. So the limiting factor on making the blades longer and skinnier is torisional rigidity. This is why carbon heli blades are popular, they have extremely high torsional rigidity. With blades with weak torsional rigidity there are problems. First is the blades unload and you lose lift. The other worse problem is that the blade twists against the springy torsional stiffness and bounces back and gets into an oscillation, flutter. The flutter creates all kinds of problems and can quickly destroy the blades.
(On G3PO, etc. The type and application of the blade covering is designed to provide a smooth efficient surface and also to provide additional torsional stiffness. This is why I include the covering and discourage the use of monokote or just paint on the blades. The aspect ratio is 10.6 : 1 and is about as far as I can push it and have the blades stay stable. The model can take additional blade length but the blades would have to be fiberglassed for additional torsional stiffness.)
If you look carefully on full sized gyros and heli's you will see airfoils with reflex at the trailing edge. This is in effect supplying trim to counter the nose down pitching moment allowing a longer more efficient blade design. The same blade without reflex would twist under its own moment.
You give up something in airfoil efficiency to get overall blade efficiency but this is the nature of having to compromise on an aircraft design.

regards
mickey
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Old Jan 22, 2007, 05:47 AM
iPhly R/C with iPhone
iter's Avatar
Silicon Valley, Calif
Joined Jan 2005
1,724 Posts
Hm. More scaling effects. Slow-G blades' aspect ratio is 12.5 : 1, and they seem to be doing fine in twist. Small is beautiful :=)

Ari.
iter is online now Find More Posts by iter
Reply With Quote
Old Jan 22, 2007, 09:36 AM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
Hop on my tractor....

About layouts. Tractor vs pusher.
The diagram shows the basics on a tractor vs pusher layout. Because the rotor has to lean back to autorotate a pusher will always have to have the motor/prop relatively lower in the airframe. The motor/prop dictates ground clearance. Because there has to be some structure below the prop to get back to the rudder, etc. this dictates even more height. The bottom line is that the pusher will always be taller than the tractor layout. Further the centerline of the thrust will always be further from the rotor drag in a pusher, with equal prop clearance from the rotor.
Basic physics says an object, unless constrained, will rotate around its center of mass. So you usually compute all the moments around the center of mass. The big items to look at here are the thrust of the motor and the drag of the rotor.
The rotor is likely to always be higher than the center of mass. So a change in rotor drag is going to pitch the aircraft until a new trim position is found (usually the pilot does something).
The thrust creates its' own moment around the center of mass. In a tractor, generally the thrust is closer to the center of mass than the tractor. Thus the distance that thrust has to act with is smaller. Drawn is the thrust below the CG in black. Adding thrust creates a nose up moment. But it's easy enough in the tractor to put downthrust in in varying amounts. It's possible to create no nose up moment with thrust or even nose down moment with thrust. Note that this doesn't get rid of the total nose up moment caused by the rotor, because as you add speed the rotor drag increases and causes a nose up moment.
But lets be careful here. There are two stages of thrust application. There is the instantaneous thrust effect and then the steady state thrust effect. Immediate thrust changes make the model pitch based on the thrust and the mass of the model because the speed hasn't changed yet, as the model changes speed the change in rotor drag comes into play and the model pitches accordingly. This describes a dynamic versus static stability problem. In the beginning the problem is dynamic, thrust around the center of mass, later the problem is static, the relative relationship of thrust and drag determining the trim.
I think ideally you want the thrust to intersect the center of mass so you don't have dynamic pitch changes with throttle. You still have to trim the rotor, but you don't get sudden pitch changes with motor application.(I originally had downthrust in G3PO and it stayed more trimmed with power. However I took it out because I prefer that the sudden burst of power get the nose up to get you out of a tight spot.)
In the pusher case it gets tougher because of the height. Generally the thrust line ends up above the CG (here , upthrust can be used to get the thrustline close to the CG). In the dynamic case added thrust creates a nose down moment. This can be a problem with full sized gyrocopters because you get into a situation with the rotor lightly loaded and you add power quickly to get out of trouble. The aircraft rolls nose down, further unloading the rotor and .... So the dynamic problem can cause a crash.
The trend in full sized aircraft is to get the heavy things (people) higher in the aircraft and the motor lower to combat this dynamic pitch problem with application of power. Models would likely benefit from the same layout.
Models can have the same problem. It can be worse because relatively speaking model props are bigger in scale than full sized props. This makes the model even taller and the CG lower. A further problem with a tall model is inertia of the body swinging. The tall, low CG model, can have the body continue to swing after power or control input, requiring more correction.
The bottom line is that the pusher takes more attention in design to minimize the height related problems, benefits from upthrust rather than downthrust and takes more attention in flight to avoid the low speed, unloaded rotor, need for high power situation.

The primary reason for a pusher layout, pilot visibility, is irrelevant in a model, so the best flying sport model is likely a tractor. But because of the popularity of pushers in the full sized world, the pusher just looks "right", so many modelers seem to head that direction. I personally did a tractor first so I had the best possible chance of learning to design and fly at the same time. The pusher came second after my piloting skills had improved and I had figured out the basic design parameters.
regards
mickey
mnowell129 is offline Find More Posts by mnowell129
Last edited by mnowell129; Jan 22, 2007 at 09:45 AM.
Reply With Quote
Old Jan 22, 2007, 05:35 PM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
What's the next topic guys?
Underslung heads?
Flybars?
Bell hiller mixing?
Twisting blades for improved performance?

mick
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Old Jan 22, 2007, 05:37 PM
iPhly R/C with iPhone
iter's Avatar
Silicon Valley, Calif
Joined Jan 2005
1,724 Posts
Twisting blades! I remember JoJo Chaulet used "broken" blades on some of his designs, but I haven't seen that used elsewhere.

Ari
iter is online now Find More Posts by iter
Reply With Quote
Old Jan 22, 2007, 05:51 PM
Winging it >
leadfeather's Avatar
Joined May 2006
9,766 Posts
It's all good, my vote is for blade twist.

Dan
leadfeather is offline Find More Posts by leadfeather
Reply With Quote
Old Jan 23, 2007, 08:09 AM
13brv3's Avatar
Navarre, FL
Joined Mar 2002
3,845 Posts
Mickey,

Thanks for the tractor vs pusher installment. I'm planning to build a full scale gyro this year, and would love to see more full size tractor options. It just seems like they have so many advantages, but as you mentioned, pilot visibility isn't one of them.

I'm still enjoying your public service series.

Thanks,
Rusty
13brv3 is offline Find More Posts by 13brv3
Site Sponsor
Reply With Quote
Old Jan 23, 2007, 08:52 AM
Mickey from Orlando. Really.
Joined Nov 2004
3,556 Posts
Quote:
Originally Posted by 13brv3
Mickey,

Thanks for the tractor vs pusher installment. I'm planning to build a full scale gyro this year, and would love to see more full size tractor options. It just seems like they have so many advantages, but as you mentioned, pilot visibility isn't one of them.

I'm still enjoying your public service series.

Thanks,
Rusty
You're welcome. Pilot visibility isn't all it's cracked up to be, especially if the view is of the rapidly approaching ground.
mnowell129 is offline Find More Posts by mnowell129
Reply With Quote
Reply


Thread Tools