Jan 31, 2013, 11:46 AM Registered User Germany, BW, Stuttgart Joined Mar 2012 803 Posts Because of where the yaw string is placed on most gliders, having the yaw string centered in a small-radius turn means that the vertical tail will "see" a sideslip angle where the relative wind is blowing from the ouside of the turn to the inside. This means that the rudder needs to be deflected to the inside of the turn in order keep the yaw string centered (the glider wants to slip due to the difference in sideslip seen by the vertical tail). Just because most gliders will show a "pedals-free" slip doesn't necessarily mean the slip is due to higher drag on the outboard wing.
Jan 31, 2013, 11:51 AM
Registered User
Germany, BW, Stuttgart
Joined Mar 2012
803 Posts
Quote:
 Originally Posted by aeronaut999 PS Shoe re your equations above, the difference in tip speed may get smaller as speed increases, but since lift scales according to airspeed squared, does the difference in lift get smaller as airspeed increases? I have always guessed the answer was "yes" but have never been completely sure... Steve
I believe the rolling moment due to to the lift difference (if not compensated for by aileron deflection) will remain constant with airpseed, but the rolling moment coefficient will be proportional to:

span*g*sin(bank_angle)/true_air_speed^2

So any need to "high side the stick" will go down like 1/true_air_speed^2
Jan 31, 2013, 11:53 AM
Ascended Master
Palmdale, CA
Joined Oct 2000
13,386 Posts
Quote:
 Originally Posted by aeronaut999 ... An aircraft that tended to skid rather than slip in a constant-banked turn would experience some very "interesting" effects-- anhedral would create a rolling-out torque, and dihedral would create a rolling-in torque! Steve
.
Yes it does.
The rudder works backwards on this one! Very difficult to fly, and nearly impossible with the wing upside down AND backwards...

# Images

Jan 31, 2013, 01:00 PM
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,165 Posts
Quote:
 Originally Posted by Sparky Paul . Yes it does. The rudder works backwards on this one! Very difficult to fly, and nearly impossible with the wing upside down AND backwards...
Ha ha!

Well-- sure the rudder works backwards, but I bet things are still "conventional" in the sense that, if you are banked and aren't making a rudder input, the anhedral tends to make a rolling-in torque. Because the aircraft is tending to slip. Not skid, which would interact with anhedral to make a rolling-out torque.

Umm.. pray tell WHY put the wing on backward as well as upside-down?

Steve
Jan 31, 2013, 01:12 PM
Ascended Master
Palmdale, CA
Joined Oct 2000
13,386 Posts
Why? Because I could!
The wing on backwards and upright flew fine. Upside down was a bit more difficult, and upside down and backwards was a single flight phenomenon!
The K-F wing kinda proved what was pretty much self evident... a plane with that wing section will have more drag than a conventional airfoil... which is prolly why no one uses it.

# Images

Jan 31, 2013, 01:13 PM
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,165 Posts
span and length and curvature

Quote:
 Originally Posted by ShoeDLG Because of where the yaw string is placed on most gliders, having the yaw string centered in a small-radius turn means that the vertical tail will "see" a sideslip angle where the relative wind is blowing from the ouside of the turn to the inside. This means that the rudder needs to be deflected to the inside of the turn in order keep the yaw string centered (the glider wants to slip due to the difference in sideslip seen by the vertical tail). Just because most gliders will show a "pedals-free" slip doesn't necessarily mean the slip is due to higher drag on the outboard wing.
Yes the vertical tail certainly plays a role. With vertical tail streamlined to the flow, we would still see some slip at the nose (and at the wing) even if there were no drag difference between the wingtips. If the outside wingtip is creating more drag than the inside wing, then the flow will impact the inside or low side of the vertical fin, with the rudder centered, and we will see a larger slip angle at the nose than if both wings were creating the same amount of drag. I suspect this is what we see in reality--and we could check it with a yaw string on the vertical fin-- but I agree that it would be theoretically possible to have the yaw string on the nose showing a slip, with the rudder centered, even if the inside wing were creating more drag than the outside wing, and the airflow were striking the outside or high side of the vertical fin at some small angle.

With my hang glider, the span is large compared to the length, so I'm pretty sure that the slip I see in the yaw string at the nose, can only be caused by the outside wing creating more drag than the inside wing, not by the way the back end of the glider "feels" a different airflow angle than the front end.

Interesting subjects to puzzle over...

Steve
 Jan 31, 2013, 01:47 PM Registered User Germany, BW, Stuttgart Joined Mar 2012 803 Posts Min sink airspeed for a glider is slower than the airspeed for L/D_max. This means that if you reduce the glider's airspeed below min sink in level flight, its total drag would go up, not down. Slowing down the inboard wing in a level turn forces it to generate the same lift at a lower speed. The inboard wing also has to mirror the rolling moment of the outboard wing (barring a lateral CG shift), and the outboard wing is seeing higher local tip speeds. This means the inboard aileron needs to be deflected downward which creates additional profile and induced drag. I can't see how a steady turn at any speed slower than the speed for L/D_max could result in a decrease in the drag on the inboard wing (or an increase in drag on the outboard wing). Also I don't think the yawing moment produced by the sideslip "seen" by the tail in a steady turn is related to the ratio of a glider's span to length... it should only be a function of the ratio of the turn radius to the displacement of the vertical tail from the CG. Last edited by ShoeDLG; Jan 31, 2013 at 02:34 PM.
 Jan 31, 2013, 04:05 PM Registered User The Willamette Valley, Oregon Joined Dec 2008 1,165 Posts Well, sounds like were going to have to film a yaw string mounted on the vertical fin of a glider, to see if it streams inboard or outboard in a no-rudder constant-banked turn. At various airspeeds. I would predict outboard, especially at low airspeeds. Just a prediction. Roughly speaking, in such a situation if the yaw string on the vertical fin streams outboard, doesn't that mean that the outboard wing is making more drag than the inboard wing? And if the yaw string streams inboard, the reverse? If both wings are making the same amount of drag, won't the vertical fin tend to streamline itself with the local flow (which would cause a yaw string at the nose to stream outboard.) The way I see it, if the span is long, the drag difference between the wing tips will tend to be larger and also will act at a much greater moment-arm from the CG, so you'll have lots of yaw torque swinging the nose toward the outside of the turn. Meanwhile if the fuselage is short, then the tail is not so far aft that there is lots of curvature in the flight path/ relative wind between the CG and the tail. All of these factors would seem to me to work together to promote an orientation where the airflow is striking the inside / low side of the vertical fin, not the outside/ high side, in a constant-bank no-rudder turn. If I'm wrong about the drag imbalance between the two tips then my picture of things may be way off. Maybe Blaine Beron-Rawdon's articles on tail size, dihedral, and spiral stability / instability are relevant here.. On my hang glider, a yaw string back at end of the "keel" definitely streams outboard in a constant-banked turn. Even in very low turns (near min sink airspeed). Required roll input is minimal-- typically a slight roll input toward the high wingtip. Emphasis on slight. The roll input involves both a CG shift and an aileron-like deformation of the trailing edges. Constant-bank turns at higher airspeed (lower angle-of-attack) require a roll input toward the low wingtip, presumably due in part to a higher "roll damping effect" due to a higher sink rate through the airmass, as well as a reduced "effective span" due to washout acting to unload the wingtips, but the direction of yaw string deflection both at nose and rear of keel is still the same. Granted the shape of the hang glider is substantially different than a conventional sailplane... This idea that if you are flying at min sink, slower than best L/D, then the inboard (slower) wing is making MORE drag than the outboard (faster) wing-- I'm not so sure. Certainly that is true when you slow down by increasing angle-of-attack, and speed up by decreasing angle-of-attack. But-- let's take the case of powered level flight-- in the turn, both wings are at the same angle-of-attack, but one is moving faster than the other. Now deflect an aileron downward on the slow wing and deflect an aileron upward on the fast wing, to equalize lift and roll torque. Maybe you end up with much more total drag on the faster wing with the raised aileron, than if you had simply pitched the whole airfoil down to a lower angle-of-attack. And maybe the same also on the slower wing with the lowered aileron-- but maybe you still end up with more drag on the faster-moving wing? I don't know... All these details aside, I think we agree that in the absence of rudder input, the majority of an aircraft's wing area will feel a slipping flow (yaw strings /telltales deflect toward high wingtip) in a constant-banked no-rudder turn in nearly all cases? Again, if not, we'll no longer get the expected rolling-out torque from any dihedral that may be present... But I have to admit that I don't have feel I have a complete grasp on the cause of this slip in all situations. I'm sure that you can't simply attribute the slip to the fact there is a net horizontal force on the aircraft. Since the aircraft is free to rotate in the yaw axis, a net horizontal force can curve the flight path without making any slip. Steve Last edited by aeronaut999; Jan 31, 2013 at 06:22 PM.
Feb 01, 2013, 05:05 AM
Registered User
Germany, BW, Stuttgart
Joined Mar 2012
803 Posts
Quote:
 Originally Posted by aeronaut999 This idea that if you are flying at min sink, slower than best L/D, then the inboard (slower) wing is making MORE drag than the outboard (faster) wing-- I'm not so sure. Certainly that is true when you slow down by increasing angle-of-attack, and speed up by decreasing angle-of-attack.
Suppose (for the sake of argument) that you have a wing with an untwisted elliptical planform and a lift distribution that is close to elliptical at all angles of attack.

If the wing is operating at a speed below the airspeed for L/D_max, and you slow it down but keep the lift constant by increasing angle-of-attack, then the wing's total drag will go up because the increase in induced drag is greater than the reduction in profile drag (subject to minor caveats).

You could also keep the lift constant (along the semispan) at reduced airspeed by deflecting an aileron downward. Will the addtional drag that results from keeping the lift constant by deflecting an aileron be more or less than the additional drag that results from keeping lift constant by increasing the wing's angle-of-attack?

If the aileron deflection kept the lift distribution elliptical, then the increase in induced drag should be exactly the same in both cases. If the aileron deflection made the lift distribution anything but elliptical, then the induced drag should increase more in the aileron deflection case.

The only way deflecting an aileron to maintain constant lift at reduced speed could reduce the drag acting on the wing semispan is if:

1. Deflecting the aileron improved the lift distribution of the wing (made it more elliptical) to the extent that the improvement in lift distribution overcame the increase in drag due to the increase in the semispan CL. This would only be the case for a wing that was performing very poorly prior to deflecting the aileron.

-or-

2. Deflecting the aileron to achieve the required increase in CL resulted in a reduction of the profile drag compared to increasing angle-of-attack to achieve the same increase in CL (again, unlikely unless the wing was performing poorly prior to deflection), and the associated reduction in profile drag coefficient was enough to overcome the the increase in induced drag (even more unlikely).

It's very unlikely a glider flying below the airspeed for L/D_max would experience a reduction in the drag on the inboard wing and increase in the drag on the outboard wing in a level turn (again, unless its performance was exceptionally poor before entering the turn).

A hang glider is slightly different. If you shift the CG laterally to maintain roll attitude rather than deflecting an aileron down, you are increasing the lift requirement (and induced drag) on the outboard wing, and reducing the lift requirement (and induced drag) on the inboard wing. It seems far more likely that a hang glider would see an overall reduction in drag on the inboard wing and an increase in the drag on the outboard wing in a level turn.

Someone with AVL skills could sort this out quickly and definitively.
Last edited by ShoeDLG; Feb 01, 2013 at 02:32 PM. Reason: spelling correction
Feb 02, 2013, 09:22 AM
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,165 Posts

Interesting thoughts Shoe...

* For the sake of talking about sideslip, the main point of interest is whether the outboard wing creates more drag than the inside wing. I didn't mean to suggest that the inboard wing necessarily saw a net drag reduction compared to in wings-level flight at the same angle-of-attack.

* Does it change anything if we say we are only interested in the part of the flight envelope that is equal to or faster than the min-sink-airspeed for the aircraft as a whole for the current bank angle, which--due to differences in lift and angle-of-attack across the span-- does not necessarily involve precisely the same angle-of-attack as measured at the wing root that would deliver min-sink in wings-level flight? In other words we likely don't want to slow the inboard wing down too much, even if that's what would be required to put the wing root at precisely the same angle-of-attack as would deliver the min-sink condition in wings-level flight...

* Stepping back to take a more holistic look at things-- consider this aircraft "ATOS VR" or "ATOS VQ"
(photos are now uploaded, see end of post)
http://www.a-i-r.de/med/nws/photos/A...0_20071106.jpg
http://www.a-i-r-usa.com/blob.asp?I=3073
http://www.a-i-r-usa.com/content.asp?ID=4339

The span is very long compared to the nose-to-tail length. There is substantial dihedral. Roll control is via spoilers ("spoilerons"). Side-to-side weight shift is minimal because with this glider, the triangular control frame is not a structural element, and when the pilot pushes sideways on the triangular control frame, the frame moves (activating the spoilers) and the pilot stays mostly fixed near the glider centerline.

Now, I'll bet you anything that the amount of dihedral was selected to give a good amount of rolling-out torque in a constant-bank turn, to balance the rolling-in tendency from the faster-moving outboard wingtip, so that the net roll torque is near zero with neither the left spoiler open nor the right spoiler open. You wouldn't want to be thermalling with a spoiler open most of the time.

I have flown one of the older ATOS gliders (like the shorter-spanned ATOS V in the last uploaded photo) and it seemed that in a constant-banked thermalling turn, no spoiler deflection was needed in either direction.

The dihedral is only going to give the expected rolling-out torque if the wing as a whole is flying in a slip (experiencing an airflow from the low wingtip toward the high wingtip). And considering the short distance from nose to tail, if a yaw string on the wing is blowing outboard, it seems likely that a yaw string on the tail also is blowing outboard. There isn't room for the flight path / airflow to curve much between the wing and the tail. Or to put it another way, the yaw rotation of the aircraft during the turn doesn't create a very large difference in yaw string angle between the CG of the aircraft and the tail.

If a yaw string on the tail is blowing outboard, then the tail is not the cause of the slipping flow over the wing. We might see even a larger slipping angle over the wing if we removed the tail.

(Many of these types of aircraft are designed to fly just fine whether the tail is installed or removed. The tail mainly provides resistance to a violent nose-down "tuck" or "tumble" in turbulent air. Certainly that is true of the shorter-span versions such as this ATOS V http://www.a-i-r-usa.com/content.asp?ID=4259 .(Last uploaded photo in this post.) Note also the absence of the vertical tip fins or winglets on the ATOS V-- the vertical tip fins or winglets are not critical to the yaw stability of these aircraft.)

I know some friends flying with these types of aircraft and could probably ask one to fly with a yaw string on nose and tail, and video cameras filming them, to confirm some of the suggestions above.

I guess it is mainly based on exposure to aircraft like this, that gives me a strong intuition than a long-spanned aircraft with a short tail moment arm will likely experiencing a slipping flow over the entire aircraft, including the tail. Due to the fact that the outboard wingtip tends to create more drag than the inboard wingtip, when net roll torque is zero. Again I can't fully explain exactly why this should be so, but I believe that it is.

Now, in the ATOS, because of sweep, drag ends up between the same between both wings in a turn, once the nose has yawed outboard to some given slip angle. If we removed the sweep, the aircraft would be yaw-unstable without a vertical fin. If we kept the fin at the same short moment-arm as the moment arm of the small V-tail in the aircraft in the photos, it would need to be very large to prevent the nose from yawing outboard to some absurd slip angle in circling flight. The slipping flow hitting the inboard side of the fin would be the thing that offsets the increased drag from the outboard wingtip. The bigger the fin, the smaller the slip angle.

If the fin were much further aft on a much longer tail boom things could be quite different-- the curvature of the flight path and airflow between the CG and the fin could possibly, at first glance, seem likely to cause the airflow to strike the outside or high side of the fin. Yet-- even here-- it seems that if both wings are making the same amount of drag, then the center of lateral area (actually laying some ways ahead of the vertical fin) will tend to position itself streamlined (tangent) to the curving flight path so that yaw strings on the aft part of the aircraft are nearly centered. If yaw strings at the rear portion of the aircraft are blowing toward the inside/ low side of the turn, this would seem to suggest that the inboard wing is creating more drag than the outboard wing, so that the fin must contribute a yawing-out torque, and if yaw strings at the rear portion of the aircraft are blowing toward the outside/ high side of the turn, this would seem to suggest that the outboard wing is creating more drag than the inboard wing, so that the fin must contribute a yawing-in torque.

I do recognize that with a long tail boom, it is possible for the rear portion of the aircraft to be nearly streamlined to the flight path / relative wind, or even for a yaw string at the tail to blow slightly inboard, even if there is a substantial slipping angle at the wing (yaw strings at wings blowing outboard). So, a slipping airflow at the rear of the aircraft is not an absolute pre-requisite for a slipping airflow over the wing. So, the fact that we generally see that dihedral creates a rolling-out torque, indicating a slipping airflow over the wing, is not absolute proof in all cases that there is a slipping airflow at the tail as well, or that the outboard wingtip is making more drag than the inboard wingtip. But my intuition is that this is the usually the case.

I intend to film a yaw string on the tail of my 2-meter Spider http://www.soaringusa.com/SPIDER-2M.html to see how it is oriented in a constant-bank turn. Noting the required aileron deflection at the same time. Which to the best of my recollection is often very near zero in a thermalling turn.

A couple more thoughts--

* If only the front portion of the ATOS gliders pictured below experienced a slipping airflow in a constant-banked turn, not the rear portions, then the increased dihedral of the outboard and rearmost wing panels (last 3 rib bays or so) would serve little purpose, at least in terms of the aerodynamics of a constant-banked turn.

* In general, don't we find that increasing the size of an aircraft's vertical fin tends to decrease sideslip while turning, which can lead to an increase in spiral instability (because dihedral now is less effective)? To me this suggests in a no-rudder turn, the airflow usually strikes the inside/ low side of the vertical fin, not the outside/ high side. Which suggests that the outboard wingtip is creating more drag than the inboard wingtip. Granted, if the bank angle and turn rate are increasing not constant, we have a demand for an increasing rate of yaw rotation, which means that yaw rotational inertia tends to promote sideslip even if the wingtips are making the same amount of drag. Also, in a case of spiral instability, I guess we aren't constraining the left and right wings to create the same amount of lift. More appropriate would be to ask, "In an aircraft with dihedral, does increasing the size of the vertical fin tend to mean that we tend to need to give more rolling-out input to keep the bank angle constant while circling at low airspeed with rudder centered?" If so, this suggests that the fin is acting to limit sideslip, suggesting that the airflow is striking the inside or low side for the vertical fin (yaw string on fin streams to outside), suggesting that the outboard wingtip is creating more drag than the inboard wingtip.

Interesting stuff...

Steve

# Images

Last edited by aeronaut999; Feb 02, 2013 at 10:43 AM.
 Feb 02, 2013, 11:21 AM Registered User The Willamette Valley, Oregon Joined Dec 2008 1,165 Posts another thought Another thought-- in general, doesn't the wingtip tend to operate at a lower l/d ratio than the wing root? And doesn't this intrisically tend to mean that the outboard wingtip tends to make more drag than the inboard wingtip in a constant-banked circle? To take an extreme case, imagine that the wing spar were extended to protude a half-span beyond each wingtip. All drag, no lift. The faster-moving extension on the outboard tip will make more drag than the slower-moving extension on the inboard tip. Isn't a real-world wing just a lesser example of this extreme case? Particularly if we have some washout. It seems that it would take a pretty idealized wing to not have more drag on the outboard wingtip than the inboard wingtip, in a constant-banked turn. Just some intuitive thoughts not a full technical explanation... steve
 Feb 02, 2013, 03:35 PM Registered User United States, UT, Salt Lake City Joined Oct 2007 6,883 Posts more drag & more lift - same thing - all depends on how you use it - -it is moving faster -
 Feb 03, 2013, 10:11 AM Have you seen my Cobra? Thousand Oaks, CA Joined Mar 2004 2,576 Posts Aero, there's no hand waving going on. Every textbook explains the cause of sideslip accurately and it's very simple: The weight vector is always vertical and lift is always perpendicular to the wing. Therefore any bank results in an uncompensated span wise component of lift. From this you get slip, and from slip you get fuselage weathervaning. The wikipedia link depicts this clearly, and though the other links apply the vector math in a confusing manner to the weight vector, they are nonetheless correct. The other causes you describe (roll rate, yaw rate) exist in reality as well as every good text but are only a miniscule and typically ignored component.
Feb 03, 2013, 10:30 AM
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,165 Posts
Quote:
 Originally Posted by vespa Aero, there's no hand waving going on. Every textbook explains the cause of sideslip accurately and it's very simple: The weight vector is always vertical and lift is always perpendicular to the wing. Therefore any bank results in an uncompensated span wise component of lift. From this you get slip, and from slip you get fuselage weathervaning. The wikipedia link depicts this clearly, and though the other links apply the vector math in a confusing manner to the weight vector, they are nonetheless correct. The other causes you describe (roll rate, yaw rate) exist in reality as well as every good text but are only a miniscule and typically ignored component.
Vespa, I think you and I are just going to have to agree to disagree. Just because there's a net force that acts horizontally in banked, turning flight, doesn't mean that the aircraft tends to be dragged sideways through the air. There's no causal link there at all. The thing is, your line of reasoning ends up predicting that the net yaw torque is not zero in a steady, constant-banked, constant-rate turn. Because you have this constant "weathervane"-based yaw torque going on even in the absence of other aerodynamic effects. It doesn't add up at all. But forget the nuances of aerodynamics and consider again the "unconventional" kinds of turns I outlined in post #35 http://www.rcgroups.com/forums/showp...2&postcount=35 . What you and are disagreeing on here is not aerodynamics but rather the basic physics of moving bodies. Steve
Last edited by aeronaut999; Feb 05, 2013 at 08:27 AM.