Originally Posted by vespa

aero, for well designed planes and pilots, the dominant contributor to adverse yaw is typically lift vector rotation. There are several other contributors (e.g. induced drag, profile drag, separation, etc.) and these are significant and often dominant, just not usually. The sideslip maneuver you describe is a perfect example of uncoordination due to the aileron deflection because there is no roll rate and thus no lift vector rotation, and this I believe is what the OP is experiencing.

I'm not quite sure what your are getting at Vespa, it sounds like circular reasoning to me. What do you mean "and pilots"? It seems to me that a good test for adverse yaw would involve an aileron input with no rudder input. If you going to give some other test for adverse yaw that involves a "well designed pilot" who always uses the rudder as intended to keep the yaw string or slip/skid ball perfectly centered, then, well, there will NEVER be any adverse yaw (unless the rudder is hitting the stops and still can't eliminate the adverse yaw, which seems most unlikely), so what would be the point of this sort of test for adverse yaw?

On a related note, a year or two ago when reading a description of Navy flight tests of roll rate of the Hellcat and Corsair, I was surprised to read that they didn't coordinate the aileron input with a rudder input. This sort of test would be interesting from an engineering/ aerodynamics standpoint but missing the point from a practical standpoint-- in the real world wouldn't a pilot coordinate his roll input with a rudder input to maximize the roll rate and eliminate the degradation of roll rate that would otherwise be caused by adverse yaw?

On the other hand when we are intentionally exploring adverse yaw we are in a different realm and clearly not a practical one-- of course no pilot would intentionally allow the aircraft to adverse-yaw under normal circumstances-- so the pilot shouldn't be a factor, we are talking about how the plane responds to aileron-only inputs, seems to me.

My "interesting" experiences in the Challenger ultralight were most decidedly based on intentional, and non-practical, aileron-only control inputs!

Steve

Originally Posted by vespa

There are several other contributors (e.g. induced drag, profile drag, separation, etc.) and these are significant and often dominant, just not usually.

I can live with that-- thanks for the note
Steve

On my model 21% of the fuselage side area is ahead of the CG and 79% is behind the CG. This includes the vertical tail area. Since the majority of the side area is behind the CG it seems like the fuselage would act like a weather vane and try to align with the relative wind when banking.

I am not sure how the wind hitting the side of the fuselage would hinder turning. it actually seems a larger side area would help turn the plane.

The larger side area does show stronger weather vaneing tendencies during takeoffs and landings vs a slim fuselage.

Originally Posted by aeronaut999

Pilot giving hard right aileron, plane rolls to a modest right bank angle but also adverse-yaws severely to the left, so that the airflow hits the right side of the fuselage creating a left sideforce that totally negates the right-pointing horizontal force vector from the right bank. Net horizontal force or turning force is zero and as pilot continues to hold strong right aileron, the plane flies in a straight line, constant heading, slight right bank, heavy left yaw angle (i.e. a yaw string streams severely to the left). The bank angle remains constant despite the pilot's heavy right aileron input, because the sideways airflow is interacting with dihedral (or the "effective dihedral" conferred by a high wing placement) to create a strong left roll torque, so the net roll torque is zero.

Here roll rate is zero so the adverse yaw torque MUST be being created by the fact that the lowered left aileron is creating more drag than the raised right aileron.

It sounds like this is what the poster is describing. His earlier aircraft perhaps had much less side area so he did not experience this before?

I have personally experienced this sort of situation in the following full-scale aircraft--

Challenger Ultralight
Ka-6 sailplane
Some others I can't think of at the moment-- maybe the Libelle sailplane, I could check my notes

The side force created by the fuselage is undoubtedly a part of the equation, but from my experience only a small one. Any fuselage is a poor wing and I doubt that its side force can compete with the horizontal lift vector component even at small bank angles where side slip and thus the fuselage effect is also weak.
A narrow, rounded fuselage like a Libelle or any other typical modern (i.e. "plastic") glider only creates a weak side force during slip as anyone who ever attempted knife-edge flight in such a glider will agree.
I've also flown the full-scale Ka6 and I disagree with your observations. There is no way to stop the turn in an un-coordinated (rudder centered) roll as long as aileron is deflected. Even in the Lo 100, which has a hughe side area compared to other gliders, a side slip during roll does not affect the roll and turn rates in such a dramatic way, even for small aileron deflections.

But I agree that asymmetric aileron drag plays a strong role when the roll begins to accelerate and roll rate is still low. This is noticeable especially in gliders where the large wing span makes asymmetric aileron drag significant. At constant aileron deflection, you must first apply quite a bit of rudder to keep the string centered. Then, as roll rate builds up and the lift difference becomes less due to the different AOAs you need to ease up on the rudder, otherwise the nose will yaw into the turn. Then, at high roll rates, tilted lift vector adverse yaw becomes dominant and you need to apply more rudder again to keep the string centered.

You should not think of it as "aileron drag" as such. The wing producing more lift will also cause more drag than the one producing less lift. This means that there is adverse yaw even in rudder only planes, but by their own nature the rudder is compensating for it in the turn. This also means that a plane with neutral roll stability will not suffer from adverse yaw while in the turn, since both wings will be producing the same amount of lift and drag.*
On large planes at low speed and large angle of attack ailerons and flaperons are either replaced or complemented with spoilers, essentially to kill lift while increasing drag on the wing inside the turn.

* However this won't really ever happen, one wing is traveling marginally slower than the other in the turn etc. But on wide turns the effect might be too small to notice.

It depends.

At normal cruise type AoA there isn't a lot of difference in profile drag between the up going and down going ailerons. Of course the airfoil, the amount of aileron deflection, and the starting AoA make a difference. If you aren't near stall, most adverse yaw is caused by the rotating lift vector from the helical motion of the wings. Dr. Drela says:

"The reason why the drag contributions are small is because deflecting the ailerons causes surprisingly little lift imbalance between the left and right wings. The roll rate which ramps up in a fraction of a second after the ailerons are applied mostly cancels the aileron-caused lift imbalance. And if the left/right lift is nearly the same, the drag will also be nearly the same as long as the flow stays attached."

I've attached a plot of a NACA 2412 (Cessna 150 airfoil) that shows how little drag difference there is between 5 degree deflections up and down, even at model Re.

As a side note, the Cessna 150/152 has Frise type ailerons to prevent adverse yaw form the lift vector rotation as well.

Kevin

Originally Posted by grant31781

On my model 21% of the fuselage side area is ahead of the CG and 79% is behind the CG. This includes the vertical tail area. Since the majority of the side area is behind the CG it seems like the fuselage would act like a weather vane and try to align with the relative wind when banking.

I am not sure how the wind hitting the side of the fuselage would hinder turning. it actually seems a larger side area would help turn the plane.

The larger side area does show stronger weather vaneing tendencies during takeoffs and landings vs a slim fuselage.

Grant, think about an attempted right turn but the plane adverse-yaws to the left, exposing the right side of the fuselage to the flow, generating a sideforce to the left. The optimum situation for this would be a small vertical fin, and lots of side area that is centered around the CG ,not well aft of the CG. Weak directional stability or yaw stability or "weathervane effect", but lots of ability to generate sideforce in a sideways flow.

I don't know whether this is the complete answer to your situation or not. You say you have plenty of side area aft of the CG which would seem to suggest strong yaw stability or directional stability or "weathervane effect", but the behavior of the model seems to suggest otherwise.

It's a misconception to imagine that generally, lots of side area will help a model turn. That might be true in some sense if you are turning with the rudder (not ailerons) so have lots of proverse yaw (not adverse yaw). It would certainly be true if you were intentionally not allowing the wing to bank as you made your turn with the rudder (by giving some opposite aileron.) But normally we don't turn that way. Of course if the side area is in the form of a vertical tail fin well aft of the CG then it will tend to minimize adverse yaw and that will help boost the roll rate, in a model with dihedral that is being controlled primarily with the ailerons.

In purely practical terms, my guess is that any or all of the following would help with your model:

* Differential aileron travel to reduce adverse yaw
* Rudder mixed to move with ailerons to reduce adverse yaw
* Larger vertical fin to reduce adverse yaw
* Less dihedral in the wing-- a completely flat wing may still have ample "effective dihedral" if it is mounted on top of a fuselage as per a high-wing Cessna-- reducing the dihedral effect will mean that any adverse yaw that does occur will have less ability to create an unfavorable roll torque and slow or stop the roll rate.

Try some of these and let us know how it works out!

(My Zagi had lousy roll handling when I bent the wings up to give it some dihedral-- combined sweep and dihedral equals way too much "effective dihedral" so the small adverse yaw that occurs even in a Zagi really slowed the roll rate-- removing the tip fins made things even worse because less "weathervane effect" equals more adverse yaw-- even though the Zagi had relatively little projected side area (especially with the tip fins off), adverse yaw still could cause big problems by dramatically slowing the roll rate. I probably led you astray by focusing on side area, it's not that important in the grand scheme of things as far as roll response to aileron inputs is concerned, but it does play an important roll in explaining why you may also see an unusually low turn rate for the bank angle at any given moment, or zero turn rate even while banked, or a turn against the direction of bank i.e. toward the high wingtip. Due to the interaction between adverse yaw, sideslip, side area, and sideforce. If you are seeing the model refuse to turn at all even though it is banked, then the side area is playing a role, but the root of the problem is still the huge amount of adverse yaw that is apparently taking place when you give an aileron input.)

Steve

JaRaMW, a few comments re your observations--

* I'm pretty sure about what I said re the Ka6. Although it was over a year ago that I last flew a Ka6, I paid careful attention and repeated the experiment several times. A sudden stick input with no accompanying rudder input could indeed cause the heading to "freeze" even though the aircraft was banked. The yaw string was streaming to the side. However one detail I did not fill in is that I allowed the rudder to float. It was not centered. It moved in the airflow. The same re the Libelle. Holding the rudder fixed would significantly increase the directional stability and the effect I'm talking about might well vanish. Also I was light in the glider which has some effect in directional stability-- the CG was near the aft edge (but still within) of the design envelope.

* I agree that the Libelle fuselage looks pretty slippery re a sideways airflow. The Ka6, less so.

* Let me note again that in the Challenger ultralight an aileron input with rudder free-floating could lead to a slight bank accompanied by a turn toward the high wingtip. That turn is driven by sideforce.

* Please consider this thought carefully-- an aircraft that generates negligible sideforce in a sideways flow, is an aircraft that needs negligible left bank to avoid entering a right turn, when the pilot holds a steady right rudder input to keep the nose continually yawed to the right of the flight path and airflow during a sideslip for a crosswind landing (wind from from right, nose aligned with runway, ground track aligned with runway) or a sideslip to steepen the descent path in calm winds (ground track and flight path aligned with runway, but nose cocked off to right.)

Do you know any aircraft that need negligible bank to hold heading when yawed cross to the flow?

The only ones I know are a Zagi (modified to have a rudder) and a hang glider (modified to have a rudder). Now there the side area is truly very small. But a conventional sailplane does always require some degree of opposite bank to hold heading as the pilot maintains a heavy rudder deflection, to the best of my knowledge. What % of the sideforce is coming from the fixed fin, versus the fuselage, in a slippery high-performance ship, would be an interesting topic to explore further.

Along the same lines, if an aircraft generates negligible sideforce in a sideways flow, then the ball will have negligible tendency to deflect in the exected direction (opposite the yaw string deflection) when the aircraft is moving sideways through the air. In high-performance sailplanes, the ball deflection is less than we would see in a more conventional airplane at the same yaw angle (yaw string angle), but it is not negligible-- the ball remains a useful tool for when the yaw string is frozen to the canopy, etc.

(The ball measures sideforce divided by lift force. When lift force is zero (zero-G's) things get interesting (slightest sideforce slams ball full to side). When lift is negative the ball needs to be mounted upside-down, i.e. the track needs to curve the other way, so the ball doesn't get stuck in one corner or the other.)

Steve

Originally Posted by JaRaMW

The side force created by the fuselage is undoubtedly a part of the equation, but from my experience only a small one. Any fuselage is a poor wing and I doubt that its side force can compete with the horizontal lift vector component even at small bank angles where side slip and thus the fuselage effect is also weak.
A narrow, rounded fuselage like a Libelle or any other typical modern (i.e. "plastic") glider only creates a weak side force during slip as anyone who ever attempted knife-edge flight in such a glider will agree.
I've also flown the full-scale Ka6 and I disagree with your observations. There is no way to stop the turn in an un-coordinated (rudder centered) roll as long as aileron is deflected. Even in the Lo 100, which has a hughe side area compared to other gliders, a side slip during roll does not affect the roll and turn rates in such a dramatic way, even for small aileron deflections.

But I agree that asymmetric aileron drag plays a strong role when the roll begins to accelerate and roll rate is still low. This is noticeable especially in gliders where the large wing span makes asymmetric aileron drag significant. At constant aileron deflection, you must first apply quite a bit of rudder to keep the string centered. Then, as roll rate builds up and the lift difference becomes less due to the different AOAs you need to ease up on the rudder, otherwise the nose will yaw into the turn. Then, at high roll rates, tilted lift vector adverse yaw becomes dominant and you need to apply more rudder again to keep the string centered.

Interesting and worthwhile thoughts. However it is also worth pointing out that many of these factors vanish when we are speaking specifically of an aircraft that is not responding to the ailerons, i.e. is not rolling, i.e. the roll rate is zero. Steve

Originally Posted by kcaldwel

It depends.

At normal cruise type AoA there isn't a lot of difference in profile drag between the up going and down going ailerons. Of course the airfoil, the amount of aileron deflection, and the starting AoA make a difference. If you aren't near stall, most adverse yaw is caused by the rotating lift vector from the helical motion of the wings. Dr. Drela says:

"The reason why the drag contributions are small is because deflecting the ailerons causes surprisingly little lift imbalance between the left and right wings. The roll rate which ramps up in a fraction of a second after the ailerons are applied mostly cancels the aileron-caused lift imbalance. And if the left/right lift is nearly the same, the drag will also be nearly the same as long as the flow stays attached."

I've attached a plot of a NACA 2412 (Cessna 150 airfoil) that shows how little drag difference there is between 5 degree deflections up and down, even at model Re.

As a side note, the Cessna 150/152 has Frise type ailerons to prevent adverse yaw form the lift vector rotation as well.

Kevin

Steve,

my guess would be that as you implied, your observation was mainly because the rudder was allowed to float and hence was deflected in the opposite direction. This yawing torque maintained a sideslip such that the fuselage's horizontal lift opposed the bank direction. This may have been amplified by low directional stability due to aft cg.
With fixed rudder, the plane will most certainly turn in the banked direction after the initial adverse yaw motion has dampened out. This is because once a constant bank angle has been established, both wings create the same lift and consequently only a small drag difference if at all. With no drag imbalance between left and right wing and no opposite rudder, directional stability will return the fuselage into a no-sideslip-condition and thus it won't produce any more sideward lift. Note that in such an experiment it is very important to prevent the nose from dropping even slightly, otherwise the descending motion will create an additional sideslip component.
Small wing drag imbalances and consequently small sideslip angles may only create significantly less lift than the horizontal comonent of a banked wing. Even for large side slip angles it is not that much, aircraft generally turn very poor with full rudder deflection and opposite aileron to keep wings level.

So I can't see how the fuselage could create sidewards lift at constant bank angle, as you described, unless the rudder is deflected opposite to aileron deflection, which leads us to an intentional sideslip with crossed ailerons and rudder.

Do you know any aircraft that need negligible bank to hold heading when yawed cross to the flow?

I found a record of a side slip I did in an ASK21, pronouncing the low amount of bank required to counter the fuselage's side-force in this aircraft type (only to answer your question; I could as well show examples of aircraft requiring more bank). There was practically no wind, so the angle with respect to the ground is pretty much identical to the side slip angle. Whether or not one calls the bank angle neglible may be a matter of taste.

slip (0 min 23 sec)

However from my experience, obtaining nice stable side slips in model airplanes is much more difficult than in full-scale ones anyway, so this should be kept in mind when talking about model planes in particular as in the OPs question.

How about the vertical CG of the plane. With the model I have the battery on the bottom of the fuselage which is about 25% of the models weight. Since an aircraft rotates around its CG I would think moving the battery up would make a difference.



I still don't get why AET planes fly fine without a rudder. Most of the AET planes I have seen have the CG close to the wing center as most are low wings. I don't recall seeing any cub type planes setup this way.

My high wing model would not fly worth a flip if the rudder was removed.

There are plenty of high wing models that fly well with no rudder. They generally keep the c.g. high (minimal dihedral effect), airspeed high (minimal CL and thus minimal adverse yaw), and use little or no dihedral and plenty of aileron differential.

Right that plane has a shallow fuselage keeping the CG close the the wing center.

The stick planes I have built were similar and did not need rudder to turn. I get that all planes have adverse yaw but most will turn fairly well with bank and yank. My Cessna like plane will not.

I am sure the CG is well below the wing. I am going to do some experimenting to see what happens with raising the battery.

I tired a few test and did a bit more observing. Moving the battery up over the wing did not seem to make much difference at all.

This plane is extremely rudder dominate! It will roll well to but the rudder is very strong. I found that the model is easiest to turn using rudder 1st then holding aileron pressure to maintain bank angle. The turns always look much better than using aileron with rudder mixed.

The plane does not like to turn real tight. In a real tight turn the plane has a somewhat severe case of over banking tendency. A bit of opposite aileron is needed to maintain all turns or the plane will spiral dive if rudder is maintained.

The adverse yaw is not so bad at speed but when flying really slow and nose high the plane take a stick full of rudder to turn. Ailerons really screw it up! I wish i could get more differential but I am using a single servo so I get what I can and its a good bit.

The new plane I am building will have 2 aileron servos.

Thanks for the reply, JaRaMW. To my way of thinking, allowing the rudder to float just more or less removes its surface area out of the picture, so that it does not contribute to the surface area of the vertical fin, as it would if you held the rudder fixed. You are basically just seeing how the plane would respond if the rudder were not present at all. I know there are many exceptions to this, but that's my starting point for how to think about this. I don't see it as contributing an opposing yaw torque in my experiment. It is just trying to streamline itself with the airflow and not fully succeeding in some cases (because it hit the stops) in which case it is still acting like part of the vertical fin to small degree and still making some small yaw torque in the same direction as the fin, i.e. trying to swing the nose back into alignment with the flow and center the yaw string. But this yaw torque is probably small compared to the fin's yaw torque, because the rudder has angled itself to a more streamlined position. Also in the sailplane experiments I don't clearly remember that the rudder hit the stops; if not then it was truly free-floating. In the Challenger case the rudder definitely did hit the stops.

Anyway, if the fin (and sometimes the rudder, if it has hit the stops) are trying to bring the yaw string back to center, and the pilot's feet are off the rudder pedals, and the rudder is not trying to bring the yaw string away from center, yet the slip angle is constant, it seems pretty clear that the wings must be generating different amounts of drag. The down-deflected aileron must be generating substantially more drag than the up-deflected one.

Sailplanes with generous vertical fins do not show this behavior but my recollection is still that the Ka-6 and the Libelle did.

Just addressing this specific situation where roll rate is zero. I'm not disagreeing that a rolling motion makes a substantial yaw torque by virtue of the "twist" in the relative wind.

In practical terms I found friction in the rudder controls to be really bothersome in some of these experiments. In some sailplanes the rudder didn't float freely but rather had some tendency to say at least slightly deflected in whatever position it was when you took your feet off the pedals. In such cases I tried to have the rudder exactly centered when taking my feet off so that my results would be biased in favor of the rudder being centered than being biased in favor of the rudder being deflected due to friction.

Originally Posted by JaRaMW

Steve,

my guess would be that as you implied, your observation was mainly because the rudder was allowed to float and hence was deflected in the opposite direction. This yawing torque maintained a sideslip such that the fuselage's horizontal lift opposed the bank direction. This may have been amplified by low directional stability due to aft cg.
With fixed rudder, the plane will most certainly turn in the banked direction after the initial adverse yaw motion has dampened out. This is because once a constant bank angle has been established, both wings create the same lift and consequently only a small drag difference if at all. With no drag imbalance between left and right wing and no opposite rudder, directional stability will return the fuselage into a no-sideslip-condition and thus it won't produce any more sideward lift. Note that in such an experiment it is very important to prevent the nose from dropping even slightly, otherwise the descending motion will create an additional sideslip component.
Small wing drag imbalances and consequently small sideslip angles may only create significantly less lift than the horizontal comonent of a banked wing. Even for large side slip angles it is not that much, aircraft generally turn very poor with full rudder deflection and opposite aileron to keep wings level.

So I can't see how the fuselage could create sidewards lift at constant bank angle, as you described, unless the rudder is deflected opposite to aileron deflection, which leads us to an intentional sideslip with crossed ailerons and rudder.


I found a record of a side slip I did in an ASK21, pronouncing the low amount of bank required to counter the fuselage's side-force in this aircraft type (only to answer your question; I could as well show examples of aircraft requiring more bank). There was practically no wind, so the angle with respect to the ground is pretty much identical to the side slip angle. Whether or not one calls the bank angle neglible may be a matter of taste.
http://www.youtube.com/watch?v=B9B26IEwHsM

However from my experience, obtaining nice stable side slips in model airplanes is much more difficult than in full-scale ones anyway, so this should be kept in mind when talking about model planes in particular as in the OPs question.