From the FAA's Pilot's Handbook of Aeronautical Knowlege (other aeronautical and physics texts make similar statements in principle): "The rotating propeller of an airplane makes a very good gyroscope and thus has similar properties. Any time a force is applied to deflect the propeller out of its plane of rotation, the resulting force is 90° ahead of and in the direction of rotation and in the direction of application, causing a pitching moment, a yawing moment, or a combination of the two depending upon the point at which the force was applied."

Emphasis added.

A gyro is a gyro is a gyro regardless of shape. It doesn't even have to be symmetrical to exibit precession. Any rotating object can precess. Just because an airplane is more complex than a toy gyro doesn't mean it magically makes precession happen without a displacement. Just because P-Factor is acting doesn't mean precession is happening. You agreed with that above, multiflyer eloquently explained it, I explained it and it's in a lot of books that way and I demonstrated it to myself years ago with a bicycle wheel and toy gyros.

You are missing the point that the force has to result in a displacement for precession to occur. If there is an equal and opposing force, then there is no displacement. If you hold a spinning bicycle wheel with an extended axle lightly by that axle, gravity will pull it down against its own "rigidity in space". As the far end of the axle rotates downward slowly, it also moves either to your left or your right, depending on which way you spun the tire. Now if you hold the axle rigidly, you are applying a force equal and opposite to gravity, and the wheel will not precess, i.e. it will not try to move left or right.

There are two principle properties of a gyro: Rigidity in Space, and Precession. If a gyro's rigidity in space is overcome, the response is precession. They are related. Rigidity in space is not a force by itself; precession is. It's the rotating version of Newton's First Law. If you pitch an airplane up, for example, and ignore all the other turning forces, the propeller will apply a right yawing force to the airplane through the prop shaft, etc, because of precession. If this force is opposed with rudder application or other forces, the airplane will not yaw. The force is there, but it has been opposed so yaw does not occur. If on the other hand the airplane is not changing orientaion/attitude, THERE CAN BE NO PRECESSION. You can't have a reaction without an action!

The above paragraph is almost entirely gibberish caused by your misunderstanding of precession. There is some truth mixed with nonsense.

What is your point? It still works and is well understood and documented. Even a relatively small force will eventually have a noticeable effect with no opposing force. And if you have ever done power on stalls with a high powered airplane, your right leg got tired. The force is so strong at high AOA, that even in a moderate left turn, you have to hold right rudder to keep the ball centered.

Maybe the same guy that wrote that also wrote that a knife-edge loop is impossible. Actually you wrote MOST model planes, so I can probably agree with that. It's all relative--small spinning mass has a large effect on a light airplane. Try power-on flat spins left and right and see the difference.

The last sentence is meaningless unless quantified. Everyone knows Spiral Slipstream exists; no-one knows exactly how strong it is under all conditions. Countering well-established aeronautical principles with no proof, experimentation or quantification is merely a literary exercise.

Incidentally, the reaction to prop thrust is that the plane moves forward. If the plane also tries to roll left slightly due to torque (which it does), then we can infer that it MUST also be applying a right rolling motion to the air, hence Spiral Slipstream. If you acknowledge torque, then if you are a reasoning person, you MUST acknowledge that the spiral component of slipstream is in proportion to the torque force applied to the airplane through the prop shaft.

All the above discussion of course ignores the minute precession that occurs in a gyro whose plane of rotation is fixed relative to the surface of the revolving and orbiting earth/galaxy/universe/set of dimensions/multiverse/antiverse/plane of existence/etherial void/quantum iteration, etc. It also ignores time dilation effects between the observer and the spinning object, space-time distortions due to gravity, the small but existent travel time of light, and the offensive nature of halitosis.

<typos and editing content>

The answer is apparantly not as obvious to some as to others -
Nothing ever changes ----------

Hello,
are you saying that it rotated into the free flow of air of the wind? Or behind a propeller? My experiment was at the end of a plane, thus into the prop airflow...

stek 79,

I was wondering if it rotated in any strong wind or just in the propwash? It you stick it out the window of a moving car does it rotate?

For what it’s worth. The first image is with the engine stopped. The threads flutter and swing back and forth in the air stream and the camera freezes them in whatever position they are in when the exposure is taken. From what I saw first hand, the center of the thread flutter and swinging was straight back from the tape holding the thread at all test points. The aircraft is a modified Tower kit built Kaos 40 powered with an Irving 53 ABC.

Your test results may differ depending on aircraft size, etc.

Now this is what it's all about!

The left nose streamer appears to be angled in the expected direction relative to the slipstream core (prop shaft axis) a few degrees and fully extended. Check.

The left rear quarter side streamers also appear angled slightly in the direction of the expected spiral component. The streamer closest to the wing may be encountering some effect from the wing. The ones on the upper corner appear to be encountering turbulence and vibrating--not fully extended-very turbulent airflow. Check, and hooray for NACA.

The left side vertical tail streamer is angled as expected, and fully extended, obviously encountering high speed flow impinging slightly on the side of the fin. A big Check.

The right side vertical tail streamer is also angled upward slightly -- not as initially expected, but it has a kink in it and is somewhat slack and hanging away from the fin slightly, leading me to believe it is not encountering strong airflow, which is still in line with the spiral slipstream observations/theory. The effect might be slightly stronger than I expected. I think that's another Check.

The streamers on the top of the fuselage appear to be responding to turbulence, possibly from the canopy. I Dunno.

The right rear quarter streamers don't appear to be encountering high speed flow--the aft one is angled perhaps too far in the "expected" direction, and the more forward one might be in a turbulent area as it appears to be pulling slightly away from the fuselage and vibrating. This also might be due to an effect of the wing and/or canopy. The further aft streamer is a check. The more forward one is inconclusive. In total, Check.

Of the 11 streamers I counted, I think the behavior of the majority can be understood to be influenced by some spiral component of the prop slipstream. There are only four that appear to be shielded from the effect entirely due to turbulence--the four near the upper aft corners of the fuselage, and possibly another two just behind the wing on the right side of the fuselage. Since it is behaving differently from its mirror, that indicates an asymmetry. Check.

The fact that the four streamers on the sides of the fuselage are behaving differently from their "mirror images" seem to point to some asymmetry between the left and right sides of the slipstream--as expected, and this could be due to spiral slipstream, offset thrust, an outside wind component, or some asymmetry in the way the streamers are mounted (i.e. not perfectly mirrored left and right). The streamers appear to be fairly symmetrically mounted, and the offset thrust is probably small on that airplane, if it has any. The aft component of slipstream is clearly the largest influence on their behavior.

I say it strongly confirms the existence of a spiral component in the slipstream. BTW, if torque exists, which we all know it does thank to Newton, the the slipstream MUST be spiraling at least a little bit initially as it leaves the prop. Torque is weak compared to thrust, and the spiral component of slipstream is weak compared to the aftward component. If a wing drags some air forward along with it (which it does), then certainly a propeller does too, as we all know a prop is a rotating wing. It all makes sense.

Anyone else want to take a stab at an armchair analysis?

As I said “The threads flutter and swing back and forth in the air stream and the camera freezes them in whatever position they are in when the exposure is taken. From what I saw first hand, the center of the thread flutter and swinging was straight back from the tape holding the thread at all test points.” The ends of the threads moved about 1/4” to 1/2” in either direction from center, but the center of the flutter on all test points was directly behind and down stream from the attachment points. I was unable to detect any spiral air flow at any of the test points. The engine on the Kaos has “0” right “0” down thrust.

If there is spiral air flow it is very weak and invisible to the test threads I used.

"Anyone else want to take a stab at an armchair analysis? "


Bugs only collect on one side of the vertical and rudder in this particular application...[img][/img]

8178,

As I suspected. Thanks for sharing that experiment.

I have played around with my electric Typhoon some more. It has a molded, rounded fuselage.

I run the motor up and down indoors. There is no doubt that with even just minimal throttle, the threads blow straight back, and with full throttle, same thing.

I have a large 3D prop on it. Sometimes, if I just give it a blip of throttle, just enough to get the prop to twist five or six times, then I can make a thread on the top center move from hanging on the left to hanging to the right. Not every time.

I am going to play with it some more, but at this point I am thinking that the slip stream twist is very quickly overwhelmed as rpm increases and the prop begins to "fly" and move large volumes of air aft. What ever twist there is, it is not enough to visibly deviate a thread on my Typhoon at minimum or maximum throttle. Maybe it is one degree or two, the threads flutter, but it is certainly nothing obvious.

Dave_Gherardini ,

That is interesting. Bugs on one side only?

That picture does not look much like a Koas with a hot engine, or a 3D foamie that will hang on the prop.

I'm wondering if the slipstream spiral is more important on lower powered, lower wing loaded models? Is it a large geared prop?

You asked.

Aspect,
I understand your point, I have not tried it into the wind! I'll borrow the TX of my friend and put it out of my car... let's hope to not bend the antenna!!!
If I think about it, I think that into free flow the windsock would stay straight, after all our big windsocks that show wind direction seems to not rotate into the wind - why would they do, from a physical point of view? The strange thing is that it seemed that there was a particular place that could force the rotation of the windsock; in other places, (I'm always speaking behind the prop flow) it seemed to stay straight... perhaps when it was exactly on the center of the flow...


8178, really nice experiment! Could it be that the wires are too close to the surface - in the boundary layer? If so, little movement could be explained thanks to fluid viscosity...


After all guys, have you ever tried to take your plane (let's say a 3D plane, with a large prop) with 0 right thrust in your hand in vertical position @ full throttle?
With mine, a FAIR intense yaw moment there will be, towards the left... in other words, my point is: the yaw moment due to spiraling slipstream is not so weak!

Yes, you would feel some yaw in addition to roll due to torque reaction. When I have done this with small airplanes, I sense primarily roll with some yaw.

I would not have expected to see more than about 3 degrees maximum of spiral, based on NACA research. In the absence of better knowlege, I would expect higher pitch props to generate more spiral than lower pitch props due to higher induced drag.

8178, I know what you wrote. If you were looking for something like 20 or 30 degrees, I can understand why it looked like zero, but are you sure there wasn't even a slight bias of something less than 3 degrees? It might be hard to see. To really measure it accurately might require carefully placed reference markings on the airframe.

To me, the most telling streamers are the left nose, and both sides of the the vertical stabilizer. You included two photos of the right side of the tail and in both cases the streamers are relatively slack compared to the one on the left side of the vertical stabilizer. That one is definitely being influenced strongly by the airflow. It is almost perfectly straight at the moment you captured it. Was it oscillating an equal amount below a line drawn on the fin parallel to the prop shaft?

Consider again the torque roll. That maneuver is indirect proof that there must be a spiral component to the slipstream (there must be a reaction to the left roll), though not overpoweringly strong. Consider that models generally require that the ailerons extend inward to very near the fuselage in order to be able to counter the left rolling effect and hover without torque rolling.

Maybe I'll do this experiment with my 35% Giles if I can get my three Yorkie pups settled long enough. It has a 26 x 10 prop.

8178, I don't mean in any way to disparage you, I am enjoying this discussion, and thanks for sharing your findings. Have you drawn any conclusions based on your experiment or do you think it's premature?

As far as a windsock spinning rapidly, that's definitely not in unison with slipstream rotation. It doesn't rotate that fast.

steck79,

Good point about 8179's experiment. The threads are very close to the surface and in the boundry layer.

I tried tying a thread to a pin. Then I stuck it into the foam on the turtle deck of my Typhoon, so it was raised above the surface 3/4 of an inch or so.

Again, the thread flutters a lot, but on average I have to say that it appeared to deviate to the right of center, maybe two or three degrees! When it was taped to the surface, I could not see that much deviation from centerline.

I tried same thing on top of the fin. It seemed to deviate even more there, maybe three or four degrees!

Very interesting! Anyone else want to stick pins in their planes? I would be curious to hear the results.

Mesae,

I want to respond to your 2/2/06 critique of my earlier post. I don’t mind criticism and challenge, but let’s lay some ground rules. Let’s ignore the rotation of the earth, relativity theory (special or general), quantum mechanics, and alternative dimensions or universes, except in case of emergency. Let’s stick to Newtonian physics. Flawed though it is, it is still useful for predicting the behavior of everyday sized physical objects at common speeds.

First of all, let me restate my main point, which no one has commented on, so far, maybe because it is a little off the subject. Before I read this thread, I believed like most other pilots that spiraling slipstream was a significant force causing an airplane to want to yaw left under various conditions. When I stumbled onto this thread, it was already four pages long, but I liked the inquisitive tone. It made me think, and I began to question some of my previously held beliefs as to why an airplane does what it does.

Actual physical evidence for the spiraling slipstream seems to be hard to find, but I think we all agree that it must be there. Also, I am not questioning the validity of any of the common explanations of P Factor found in most flight manuals.

However, while I was pondering the earlier discussions, it just occurred to me, out of nowhere, that there could be another cause for the left yaw at positive AOA. This is not directly related to spiraling slipstream, but I thought I would toss the idea out there, and see if anyone had any thoughts on it. I am not doubting the existence of the usual suspects, but I am just proposing that there may be another force that has not been discussed, so far as I know. I am suggesting that there should be a yawing force on the airframe that is not a result of forces at the prop hub.

What I am proposing is that there should be a Bernoulli effect. Bernoulli’s law states, “When the speed of a fluid increases, the pressure in the fluid decreases.” If you blow over the top of a soda straw in a glass of liquid, the fluid level in the straw rises. The asymmetrical prop loading at positive AOA causes an asymmetrical acceleration of the air. The air should be moving faster on the right, therefore, the air pressure on that side of the fuselage should be lower, from the nose to the tail. Is there something wrong with my reasoning here? Does this make sense to anyone besides me? Maybe this belongs in another discussion.

Now, back to the gyroscopic effects, briefly, one more time, and then you and Dick Hanson can call it gibberish if you want, but LouW and other readers may want to think about it some more. A spinning propeller would act just like a gyroscope, but only in a vacuum. When you put it in air, aerodynamic forces, such as asymmetrical prop loading, can come into play. There can be no asymmetrical prop loading on a simple gyroscope, but a propeller is a device with both aerodynamic and gyroscopic properties. A gyroscope is just a gyroscope. They are not equivalent devices.

This is your quote from FAA Pilot Handbook. "The rotating propeller of an airplane makes a very good gyroscope and thus has similar properties. Any time a force is applied to deflect the propeller out of its plane of rotation, the resulting force is 90ĚŠ ahead of and in the direction of rotation and in the direction of application, causing a pitching moment, a yawing moment, or a combination of the two depending upon the point at which the force was applied."

This looks like a perfectly good definition to me. “Any time” a force is applied to deflect the plane of a gyro in one direction (such as asymmetrical prop loading would do), the result is a force in another direction. Isn’t that what it says? A gyroscope is a device that redirects force. I do not see where it says anything about motion causing force. I repeat, according to Newton, force results in motion, not the other way around. This may not be as obvious to some as it is to others, but that’s my story, and I am sticking to it.

I will admit that it is not all that obvious. There are web forums devoted to this subject. There are Ph.D.’s who know all the mathematical formulas and can do the calculations, but don’t really get it.

Forget about motion causing force. If you think about it the other way around and reread my posts, they might make more sense.

Maybe this will help. There are no perfect gyroscopes in nature, just like there are no perfectly rigid bodies, and no perfect circles. These are just concepts, but they have proven themselves to be useful. (I wax philosophical, but we did not say no metaphysics allowed.)

Enough on gyroscopes.

Let’s get back to the slipstream.

Your Bernoulli idea may have some merit.

As for precession, you are missing the point. What you are implying is that precession can happen without a displacement. If you study precession with an open mind, maybe do an experiment, you will come to a better understanding of it.

I'm not sure I understand what you keep saying about motion causing a force. You seem to be implying that that is what I'm saying. It is not. Precession is a reaction to displacement, which is caused by a force. If you don't see by the FAA's explanation that the propeller's plane of rotation has to rotate, you have much of the right idea, again, you are just missing the point.

Don't be stubborn. Study it carefully and you will be rewarded.

The fact that a propeller is moving air does not affect its gyroscopic properties. Toy gyros have an effect on air as well.

The universe stuff was a joke.

If you are suggesting that I am one of those people who has a lot of theoretical knowledge yet doesn't "get it", think again. I do have some theoretical knowledge, but I can more than back it up with 27 years of modeling experiece, full-scale aerobatic competition and 4600 hours full-scale flight time, most of it professional, and an Airline Transport Pilot Certificate. Also, four years professional UAV R&D and revenue flight experience.

OK I did it. I took my plane out to the side of the house, using fine white thread mounted several inches from the top of the cowl on a wire, with the plane facing into a light wind such that with the engine off, the streamer didn't favor one side or the other.

That was a little hard to see on photos so I took it to the most sheltered area outside my house and used red ribbon. The wind reaching the area was very light and variable. again I positioned the plane so the streamer didn't favor one side or the other before starting the engine.

I mounted the streamers on a 5" (or so) piece of wire, holding it well away from the slipstream core, because I reasoned that the effect would grow progressively stronger toward the propeller tips, since the tangent velocity of the prop and therefore the spiral is MUCH higher there, which might explain why thread taped to the side of the fuselage (or oil on the fuse) does not show the effect as conclusively.

As you can see, the spiral component is unmistakable, especially considering the engine is mounted with right thrust, which would tend to decrease the angle the streamer makes with the fuselage centerline. I even altered the position of the plane for some of the photos (the head-on picture is one) so that the light prevailing wind tended to decrease the angle (blow it back toward the fuselage), and at relatively high power settings (1/3 throttle), the streamer still spent by far most of its time on the right side of the centerline as long as I didn't angle it with too much of a crosswind. With full throttle, the crosswind might not have mattered much. The streamer did come over to the "wrong" side a few times, but only briefly each time. The streamers clearly and overwhelmingly favored the right side of the fuselage centerline. This can also be seen in the videos I shot.

You can't see it in the photo, but I was protected by a pipe while taking the head-on picture, in case the plane broke loose. I always try to be safe.

I never exceeded about 1/3 throttle because I was nervous about the stake coming out of the ground. It probably wouldn't have, but I didn't want to stress the tailwheel either, and I didn't have a spotter.

I have a few short videos too. I'll try to post them and edit this thread with links when I get that done. Sorry about the tiny, low quality videos; cable and DSL are not available in my neighborhood yet. I think I'll only post the smaller videos with the red ribbons for now. The first two vidoes with the white thread show the same result under less favorable conditions (more wind), and are much larger files.

Here are two little videos:

http://media.putfile.com/Spiral-Slipstream-1
http://media.putfile.com/Spiral-Slipstream-2

P.S. Here is due credit to aspectratio for jogging my memory about the tangent velocity of the spiral as a function of its distance from the core.

Msae,
this is the final experiment! No more question about splipstream existence.... GOOD WORK!!!

Only one question, you didn't try to give full throttle? Too bad!

Thank you again!

yep -the air has to be moving in response to the airscrew direction.
The flow pattern tho-------- has too many factors involved to see how much pressure is going BACK vs going sideways.
It is there and will be different on every plane and at different airspeed and at different power settings . That is why trim tabs and such were invented.
On a related note -- cooling our model engines is also related to the spiral output of the airscrew .
The prop does not -as some think - blow air into the cowl inlets.
It just stirs up things .
Then
The air flowing over the cowl, creates a low pressure at outlets/vents/louvers-which is replaced by air in the cowl-so the inlets of the cowl only provide a make up source for air flowing thru th cowl.
On full scale the cylinders are well ducted to really make the airflow hug the engine fins --and PULL the air thru to th outlet.
On my CAP and my Edge -I tried two other tricks -whch seem to work:
the Cap got vents added alon top the cowl - to receive air further from the hub area air and above the engine - this helped the internal flow to start above the engine -then down and out . Made an instant difference in temperatures .
On my EDGE-- I simply added a extension to the front of the cowl - to again extend the distance from the prop -to the cylinder (located at lower left of picture) and the scale inlets - the idea was the same to hopefully gain access to air not simply stirring at the hub.
Spiral flow is spiral flow and the prop twist at the hub area does really nothing to add rearward flow of air- so These approaches were done to help create some positive pressure in this area. The Cap is flowna lot at shows etc., now by a friend and is constantly staggering along at high AOA and the engine HAS to be kept cool .

Also -on the big radial engined setups - the initial flow from the prop gets altered by the proximity of the cowl to the prop -on th YAK (full scale ) the spinner sets well forward -I did the same on my model - works well
note the scale locations on the cowl for exit air (fixed louvers)

Why? Were you hoping it would break loose and hurt me?

Certainly not!

I wanted to see the stream motion at full throttle, that is the situation where we see most of the left yaw effects...

Well taken

The average angle didn't seem to increase very much between a fast idle and the max power I applied --- only the "authority" with which the streamer was extended.

I think Dick’s experience explains why the airflow on the Kaos does not exhibit spiral airflow along the fuselage area or vertical stab. I did some testing of air flow on a medium size household cooling fan yesterday (safer than playing with streamers around a running aircraft engine) and found the same thing Dick discussed. The airflow is weak and somewhat mixed up in the hub area of the blades. But strength of the spiral airflow can be observed when you move the streamer out toward the tip of the blades. With the streamer up close to the blades and the blade tip it is very pronounced and less pronounced down stream from the blade. The airflow down stream from the fan looks almost like a hollow tube, strong at the outer ring and week and mixed up in the center area.

Like many of the 60s, 70s and 80s vintage pattern aircraft the engine on the Kaos is very tightly enclosed around the crank case and tapers down to a very small spinner and has a thin fuselage section all the way back to the tail. I expect the airflow along thin fuselage is being produced by the inner part of the blades and is a weak airflow, thus it does not show any strong spiral airflow component. But think about what happens to this part of the weak airflow after the aircraft is up to speed and flying 80 MPH or faster.

I noticed a comment in an on-line aerodynamics reference for full scale aircraft that the yaw impact from spiral air flow only happens if the vertical stab is tall enough to reach into the spiral component of the airflow. Very interesting.

If you want to see the flip side of that -- get a pusher setup and holding a bit of yarn -- watch the downstream flow
when there is a large nacelle at the incoming air and no impedance on the downstream air - --I found that the air twists into a tight small stream-
On my two worthless twin boom pusher pattern designs - I was surprised at the different flow (tractor vs pusher)
Tho old Kaos and other small box fuselages had another crazy characteristic.
If you placed a big canopy - "just right" on the fuselage - the model would develop a pronounced tail wag -very cyclic.
Also on early fun fly models when doing an "egg drop" the egg cup - mounted to forward top of fuselage would create the same disturbance and the tail would search right to left - til it hit a node that made it wiggle like a hula dancer -

Very interesting guys!