Bent up wings on Schempp-Hirth and Jonkers glider

[Jonathan St. Cloud]

Can an aerodynamicists explain the reasoning behind the bent up wing tips of the Schempp-Hirth and Jonkers gliders. I thought Schempp first started this on the Nimbus4 purely to keep the outboard tips from getting scraped, but now the tips are bent on shorter wing birds. Someone fairly knowledgeable once told me the the bent tips actually hurt performance in the run but help in climb. I just wanted to get a bt more educated not the reasoning. I noticed the Quintus has bent tips but the same wing on the Antares 23 were straight.

[J. Nieuwenhuize]

If we go back to theory, the perfect wing is a double super-ellipse (Lamé curve):
http://en.wikipedia.org/wiki/Superellipse

Both the top view and the dihedral of the wing should have the shape of a super-ellipse for lowest induced drag for a given bending moment (structural weight) and wetted area (profile drag at high speeds). Both the A350 and the Dreamliner are very close to this ideal:
wallpaperswide.com/download/boeing_787_dreamliner-wallpaper-1920x1200.jpg

Another plus compared to a wing without dihedral and winglets is that the interference drag between the winglet and the wing is much reduced.

Such a gradually curved wing is impossible to build because all the control surfaces would have a bend in them. The wing with sections progressively canted more and more (polyhedral) is a good compromise.

It also helps with flutter apparantly. The Vortex shedding frequency of the various sections make the critical flutter speed for such a wing higher, allowing either a higher VNE, or a less stiff and thus lighter wing structure.

[Mike the Strike]

I don't know if it was the extra tip dihedral or some other aspect of the design, but I found the Discus 2 to be the easiest glider I've ever thermalled - like it was on rails. It wasn't shabby on the run either.

Mike

[Andy Blackburn[_3_]]

On Monday, May 18, 2015 at 1:34:15 PM UTC-7, J. Nieuwenhuize wrote:
> If we go back to theory, the perfect wing is a double super-ellipse (Lamé curve):
> http://en.wikipedia.org/wiki/Superellipse
>
> Both the top view and the dihedral of the wing should have the shape of a super-ellipse for lowest induced drag for a given bending moment (structural weight) and wetted area (profile drag at high speeds). Both the A350 and the Dreamliner are very close to this ideal:
> wallpaperswide.com/download/boeing_787_dreamliner-wallpaper-1920x1200.jpg
>
> Another plus compared to a wing without dihedral and winglets is that the interference drag between the winglet and the wing is much reduced.
>
> Such a gradually curved wing is impossible to build because all the control surfaces would have a bend in them. The wing with sections progressively canted more and more (polyhedral) is a good compromise.
>
> It also helps with flutter apparantly. The Vortex shedding frequency of the various sections make the critical flutter speed for such a wing higher, allowing either a higher VNE, or a less stiff and thus lighter wing structure.
h

Elliptical polyhedral is not part of any drag theory I ever learned studying aerodynamics. The wing planform, airfoils, twist and the use of winglets are used together to optimize the tradeoff between parasitic and induced drag while maintaining desirable handling and stall characteristics. My sense is that use of dihedral (or polyhedral) is mostly motivated by handling (and perhaps ground clearance) considerations rather than performance considerations. They may also think it looks cool.

9B

[[email protected]]

On Monday, May 18, 2015 at 3:56:47 PM UTC-4, Jonathan St. Cloud wrote:
> Can an aerodynamicists explain the reasoning behind the bent up wing tips of >the Schempp-Hirth and Jonkers gliders. I thought Schempp first started this >on the Nimbus4 purely to keep the outboard tips from getting scraped, but now >the tips are bent on shorter wing birds. Someone fairly knowledgeable once >told me the the bent tips actually hurt performance in the run but help in >climb. I just wanted to get a bt more educated not the reasoning. I noticed >the Quintus has bent tips but the same wing on the Antares 23 were straight.

Here is a theory: When you are banked at 40° when thermalling the inside
wing tip is closer to the core than the outside wingtip. With the strong dihedral of the wingtip this puts the bank angle of the wing tip closest to the center of the thermal at say ~28° instead of the ~40° so the lift vector is closer to perpendicular to the lift = better climb rate.

What i don't understand is that when flying straight/level with an effectively high dihedral the lift vector of both wings are pointing perpendicular to the wings, so the vectors are not pointing up, but rather inward. They balance each other out of course but seems like if they were perpendicular to gravity they would be most efficient. Isn't dihedral costly on performance, but helps on handling?

Curious if anyone else makes any sense of this. I'm more asking than telling.
Chris
Not an aerodynamic expert!

[J. Nieuwenhuize]

Op dinsdag 19 mei 2015 04:38:19 UTC+2 schreef Andy Blackburn:
> On Monday, May 18, 2015 at 1:34:15 PM UTC-7, J. Nieuwenhuize wrote:
> > If we go back to theory, the perfect wing is a double super-ellipse (Lamé curve):
> > http://en.wikipedia.org/wiki/Superellipse
> >
> > Both the top view and the dihedral of the wing should have the shape of a super-ellipse for lowest induced drag for a given bending moment (structural weight) and wetted area (profile drag at high speeds). Both the A350 and the Dreamliner are very close to this ideal:
> > wallpaperswide.com/download/boeing_787_dreamliner-wallpaper-1920x1200.jpg
> >
> > Another plus compared to a wing without dihedral and winglets is that the interference drag between the winglet and the wing is much reduced.
> >
> > Such a gradually curved wing is impossible to build because all the control surfaces would have a bend in them. The wing with sections progressively canted more and more (polyhedral) is a good compromise.
> >
> > It also helps with flutter apparantly. The Vortex shedding frequency of the various sections make the critical flutter speed for such a wing higher, allowing either a higher VNE, or a less stiff and thus lighter wing structure.
> h
>
> Elliptical polyhedral is not part of any drag theory I ever learned studying aerodynamics. The wing planform, airfoils, twist and the use of winglets are used together to optimize the tradeoff between parasitic and induced drag while maintaining desirable handling and stall characteristics. My sense is that use of dihedral (or polyhedral) is mostly motivated by handling (and perhaps ground clearance) considerations rather than performance considerations. They may also think it looks cool.
>
> 9B

No, it's not just a matter of handling and cool looks.

Most universities don't go further than lifting line theory. The name already gives away that is has it's issues; it's a 2D theory.
It's not for cosmetic reasons that the first generation of airliners and bizjets that can make use of new understanding and new construction methods all converge to polyhedral wings where the winglets are blended in the wing design.

[Andy Blackburn[_3_]]

On Monday, May 18, 2015 at 8:00:29 PM UTC-7, wrote:
> On Monday, May 18, 2015 at 3:56:47 PM UTC-4, Jonathan St. Cloud wrote:
> > Can an aerodynamicists explain the reasoning behind the bent up wing tips of >the Schempp-Hirth and Jonkers gliders. I thought Schempp first started this >on the Nimbus4 purely to keep the outboard tips from getting scraped, but now >the tips are bent on shorter wing birds. Someone fairly knowledgeable once >told me the the bent tips actually hurt performance in the run but help in >climb. I just wanted to get a bt more educated not the reasoning. I noticed >the Quintus has bent tips but the same wing on the Antares 23 were straight.
>
> Here is a theory: When you are banked at 40° when thermalling the inside
> wing tip is closer to the core than the outside wingtip. With the strong dihedral of the wingtip this puts the bank angle of the wing tip closest to the center of the thermal at say ~28° instead of the ~40° so the lift vector is closer to perpendicular to the lift = better climb rate.
>
> What i don't understand is that when flying straight/level with an effectively high dihedral the lift vector of both wings are pointing perpendicular to the wings, so the vectors are not pointing up, but rather inward. They balance each other out of course but seems like if they were perpendicular to gravity they would be most efficient. Isn't dihedral costly on performance, but helps on handling?
>
> Curious if anyone else makes any sense of this. I'm more asking than telling.
> Chris
> Not an aerodynamic expert!

Where to begin.

In your example the inner wing is going slower and produces slightly less lift (everything else being equal) and the outer tip is going slightly faster and produces more lift, so by your logic the extra dihedral overall would go towards producing more inward lift than upward lift.

Of course none of that really explains use of polyhedral versus dihedral. Both are used to create spiral mode stability. Simple V dihedral is easier to manufacture, whereas polyhedral will produce similar spiral stability with more of the wing carrying a lower dihedral angle overall because of the longer moment arm out at the wingtip. In either case, the dihedral effect occurs through the coupling of roll and sideslip.

It's really not first and foremost a performance thing. The angles involved are pretty small in the first place so the cosine for the dihedral angle will be close to 1 and the sine will be close to 0, meaning the performance effect will be tiny. You mention something like 12 degrees of dihedral in your example, but that much would produce a pronounced dutch roll and would be a pretty unpleasant to fly airplane. More typical is a couple of degrees..

Also keep in mind how angle of attack is produced by pitching the aircraft. The whole idea of producing lift parallel to the span requires either a fixed angle of incidence relative to the direction of the travel (such as for winglets) or lift that is created as a function of sideslip to create spiral stability (as is the case for dihedral and polyhedral). Wikipedia has a decent explanation. http://en.wikipedia.org/wiki/Dihedral_(aeronautics).

9B

[Martin Gregorie[_5_]]

On Mon, 18 May 2015 19:38:18 -0700, Andy Blackburn wrote:

> Elliptical polyhedral is not part of any drag theory I ever learned
> studying aerodynamics. The wing planform, airfoils, twist and the use of
> winglets are used together to optimize the tradeoff between parasitic
> and induced drag while maintaining desirable handling and stall
> characteristics. My sense is that use of dihedral (or polyhedral) is
> mostly motivated by handling (and perhaps ground clearance)
> considerations rather than performance considerations. They may also
> think it looks cool.
>
Elliptical polyedral and planform have been described as the ideal and
used for years in the design of high performance free flight competition
models. There are references going back to the early '60s: Jim Baguley's
articles on F1A design in Aeromodeller, several articles in the annual
NFFS Symposium reports since 1968. These suggested that approximating an
elliptical area distribution minimises tip drag, while doing the same for
polyhedral minimises the tip height and hence the total wing area for a
given projected area, with the added benefit that, because polyhedral
minimises the angle between adjacent panels, it also minimises
interference drag. Six panel wings have been common in the F1ABC classes
for the last 15-20 years.

But then, as Will Schueman said, this is to be expected since the model
design/build generation time is much shorter than that for sailplanes:
6-12 months vs 5+ years, so more rapid evolution is to be expected.


--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[[email protected]]

All those arguments seem not to consider flexibility, watch at an open class glider wings at high Cl....

Carlo

[Andy Blackburn[_3_]]

On Tuesday, May 19, 2015 at 1:32:02 AM UTC-7, Martin Gregorie wrote:
> On Mon, 18 May 2015 19:38:18 -0700, Andy Blackburn wrote:
>
> > Elliptical polyhedral is not part of any drag theory I ever learned
> > studying aerodynamics. The wing planform, airfoils, twist and the use of
> > winglets are used together to optimize the tradeoff between parasitic
> > and induced drag while maintaining desirable handling and stall
> > characteristics. My sense is that use of dihedral (or polyhedral) is
> > mostly motivated by handling (and perhaps ground clearance)
> > considerations rather than performance considerations. They may also
> > think it looks cool.
> >
> Elliptical polyedral and planform have been described as the ideal and
> used for years in the design of high performance free flight competition
> models. There are references going back to the early '60s: Jim Baguley's
> articles on F1A design in Aeromodeller, several articles in the annual
> NFFS Symposium reports since 1968. These suggested that approximating an
> elliptical area distribution minimises tip drag, while doing the same for
> polyhedral minimises the tip height and hence the total wing area for a
> given projected area, with the added benefit that, because polyhedral
> minimises the angle between adjacent panels, it also minimises
> interference drag. Six panel wings have been common in the F1ABC classes
> for the last 15-20 years.
>
> But then, as Will Schueman said, this is to be expected since the model
> design/build generation time is much shorter than that for sailplanes:
> 6-12 months vs 5+ years, so more rapid evolution is to be expected.
>
>
> --
> martin@ | Martin Gregorie
> gregorie. | Essex, UK
> org |

I have to admit to having a bit of trouble with the idea that having two steps in the polyhedral of 3 degrees each followed by a 84 degree angle for the winglet has much impact on interference drag. Gliders with span limits (for class or structural weight considerations) still generally have a vertical winglet at the tip rather than the Boeing-style flat raked tip (though a winglet and a span extension have similar effects on wingtip vortex and induced drag reduction for slightly different tradeoffs in bending moment).

I can accept the idea that raking the tip near the winglet affects spanwise flow and may have some beneficial effect on the transition. We've known about the potential benefits of sweeping the leading edge since Will Schuemann started modifying his ASW-12 and probably before that. If interference drag at the winglet junction were the big factor everyone would have LS-8-style winglets. I suspect the radius to reduce interference drag at these Reynolds numbers is measured in inches, not tens of yards.

I also get that polyhedral may give you similar handling for less wetted area than v-dihedral and that this may have become more attractive with the advent of stiffer carbon wings that don't give you dihedral through bending as much, but seriously, it has to be a fraction of a percent since we are talking about needing more polyhedral at the tip to yield similar spiral stability to low single-digit dihedral at the root. IMHO the additional tip clearance may throw enough weight in favor of the polyhedral design to make it worth the additional construction complexity.

You certainly are seeing it in multiple designs now.

9B

[[email protected]]

Who need polydiedhral?

Whatch this

http://www.aeromedia.it/lb1641.jpg

[J. Nieuwenhuize]

Op dinsdag 19 mei 2015 17:33:31 UTC+2 schreef Andy Blackburn:
> On Tuesday, May 19, 2015 at 1:32:02 AM UTC-7, Martin Gregorie wrote:
> > On Mon, 18 May 2015 19:38:18 -0700, Andy Blackburn wrote:
> >
> > > Elliptical polyhedral is not part of any drag theory I ever learned
> > > studying aerodynamics. The wing planform, airfoils, twist and the use of
> > > winglets are used together to optimize the tradeoff between parasitic
> > > and induced drag while maintaining desirable handling and stall
> > > characteristics. My sense is that use of dihedral (or polyhedral) is
> > > mostly motivated by handling (and perhaps ground clearance)
> > > considerations rather than performance considerations. They may also
> > > think it looks cool.
> > >
> > Elliptical polyedral and planform have been described as the ideal and
> > used for years in the design of high performance free flight competition
> > models. There are references going back to the early '60s: Jim Baguley's
> > articles on F1A design in Aeromodeller, several articles in the annual
> > NFFS Symposium reports since 1968. These suggested that approximating an
> > elliptical area distribution minimises tip drag, while doing the same for
> > polyhedral minimises the tip height and hence the total wing area for a
> > given projected area, with the added benefit that, because polyhedral
> > minimises the angle between adjacent panels, it also minimises
> > interference drag. Six panel wings have been common in the F1ABC classes
> > for the last 15-20 years.
> >
> > But then, as Will Schueman said, this is to be expected since the model
> > design/build generation time is much shorter than that for sailplanes:
> > 6-12 months vs 5+ years, so more rapid evolution is to be expected.
> >
> >
> > --
> > martin@ | Martin Gregorie
> > gregorie. | Essex, UK
> > org |
>
> I have to admit to having a bit of trouble with the idea that having two steps in the polyhedral of 3 degrees each followed by a 84 degree angle for the winglet has much impact on interference drag. Gliders with span limits (for class or structural weight considerations) still generally have a vertical winglet at the tip rather than the Boeing-style flat raked tip (though a winglet and a span extension have similar effects on wingtip vortex and induced drag reduction for slightly different tradeoffs in bending moment).
>
> I can accept the idea that raking the tip near the winglet affects spanwise flow and may have some beneficial effect on the transition. We've known about the potential benefits of sweeping the leading edge since Will Schuemann started modifying his ASW-12 and probably before that. If interference drag at the winglet junction were the big factor everyone would have LS-8-style winglets. I suspect the radius to reduce interference drag at these Reynolds numbers is measured in inches, not tens of yards.
>
> I also get that polyhedral may give you similar handling for less wetted area than v-dihedral and that this may have become more attractive with the advent of stiffer carbon wings that don't give you dihedral through bending as much, but seriously, it has to be a fraction of a percent since we are talking about needing more polyhedral at the tip to yield similar spiral stability to low single-digit dihedral at the root. IMHO the additional tip clearance may throw enough weight in favor of the polyhedral design to make it worth the additional construction complexity.
>
> You certainly are seeing it in multiple designs now.
>
> 9B

Comparing a planar (no dihedral) wing with a winglet to a "perfect" super-elliptic continously changing polyhedral wing we're talking on the order of a percent lower induced drag and another percent less drag due to not having to suffer interference drag. With the present polyhedral wings, you're probably down to less than half that. That's under 1% less induced drag, or about half a percent less drag at low speeds.

This is an interesting result. Having seen the work in a bit more detail, having a big radius (or a continously changing polyhedral) most certainly pays off, especially at the high Cl's we fly a considerable amount of time at:
http://www.apollocanard.com/4_blended%20winglets.htm

[Andy Blackburn[_3_]]

On Tuesday, May 19, 2015 at 12:12:49 PM UTC-7, J. Nieuwenhuize wrote:
> Op dinsdag 19 mei 2015 17:33:31 UTC+2 schreef Andy Blackburn:
> > On Tuesday, May 19, 2015 at 1:32:02 AM UTC-7, Martin Gregorie wrote:
> > > On Mon, 18 May 2015 19:38:18 -0700, Andy Blackburn wrote:
> > >
> > > > Elliptical polyhedral is not part of any drag theory I ever learned
> > > > studying aerodynamics. The wing planform, airfoils, twist and the use of
> > > > winglets are used together to optimize the tradeoff between parasitic
> > > > and induced drag while maintaining desirable handling and stall
> > > > characteristics. My sense is that use of dihedral (or polyhedral) is
> > > > mostly motivated by handling (and perhaps ground clearance)
> > > > considerations rather than performance considerations. They may also
> > > > think it looks cool.
> > > >
> > > Elliptical polyedral and planform have been described as the ideal and
> > > used for years in the design of high performance free flight competition
> > > models. There are references going back to the early '60s: Jim Baguley's
> > > articles on F1A design in Aeromodeller, several articles in the annual
> > > NFFS Symposium reports since 1968. These suggested that approximating an
> > > elliptical area distribution minimises tip drag, while doing the same for
> > > polyhedral minimises the tip height and hence the total wing area for a
> > > given projected area, with the added benefit that, because polyhedral
> > > minimises the angle between adjacent panels, it also minimises
> > > interference drag. Six panel wings have been common in the F1ABC classes
> > > for the last 15-20 years.
> > >
> > > But then, as Will Schueman said, this is to be expected since the model
> > > design/build generation time is much shorter than that for sailplanes:
> > > 6-12 months vs 5+ years, so more rapid evolution is to be expected.
> > >
> > >
> > > --
> > > martin@ | Martin Gregorie
> > > gregorie. | Essex, UK
> > > org |
> >
> > I have to admit to having a bit of trouble with the idea that having two steps in the polyhedral of 3 degrees each followed by a 84 degree angle for the winglet has much impact on interference drag. Gliders with span limits (for class or structural weight considerations) still generally have a vertical winglet at the tip rather than the Boeing-style flat raked tip (though a winglet and a span extension have similar effects on wingtip vortex and induced drag reduction for slightly different tradeoffs in bending moment).
> >
> > I can accept the idea that raking the tip near the winglet affects spanwise flow and may have some beneficial effect on the transition. We've known about the potential benefits of sweeping the leading edge since Will Schuemann started modifying his ASW-12 and probably before that. If interference drag at the winglet junction were the big factor everyone would have LS-8-style winglets. I suspect the radius to reduce interference drag at these Reynolds numbers is measured in inches, not tens of yards.
> >
> > I also get that polyhedral may give you similar handling for less wetted area than v-dihedral and that this may have become more attractive with the advent of stiffer carbon wings that don't give you dihedral through bending as much, but seriously, it has to be a fraction of a percent since we are talking about needing more polyhedral at the tip to yield similar spiral stability to low single-digit dihedral at the root. IMHO the additional tip clearance may throw enough weight in favor of the polyhedral design to make it worth the additional construction complexity.
> >
> > You certainly are seeing it in multiple designs now.
> >
> > 9B
>
> Comparing a planar (no dihedral) wing with a winglet to a "perfect" super-elliptic continously changing polyhedral wing we're talking on the order of a percent lower induced drag and another percent less drag due to not having to suffer interference drag. With the present polyhedral wings, you're probably down to less than half that. That's under 1% less induced drag, or about half a percent less drag at low speeds.
>
> This is an interesting result. Having seen the work in a bit more detail, having a big radius (or a continously changing polyhedral) most certainly pays off, especially at the high Cl's we fly a considerable amount of time at:
> http://www.apollocanard.com/4_blended%20winglets.htm

6-11" radius seems reasonable for interference drag reduction and if I look at the JS-1C 3-view it seems to have a winglet transition with around 8-10 cm radius (and a smaller chord than the apollo canard).

That doesn't explain the OP question about polyhedral. The JS-1C 3-view appears to show polyhedral breaks at 6, 7.5 and 9.5 meters with radii of on the order of 5-10 meters, not centimeters, so it would appear two 50-100 times too big to be of practical use in reducing interference drag, plus the reduction of the included angle between the wing tip and the winglet is too minor that this is the reason when you can reduce the interference drag simply by upping the transition radius to the winglet by a centimeter or two.

I remain skeptical that polyhedral in these modern wings is motivated by winglet aerodynamics or a drive for meaningful induced drag reduction. At least the arguments posted so far are not compelling on the point.

But I have a high need to verify things...

9B

[Martin Gregorie[_5_]]

On Tue, 19 May 2015 08:33:29 -0700, Andy Blackburn wrote:

> I have to admit to having a bit of trouble with the idea that having two
> steps in the polyhedral of 3 degrees each followed by a 84 degree angle
> for the winglet has much impact on interference drag.
>
They're not at all common, and don't seem to help a lot in F1ABC class
models and are not commonly used. There's have been a few F1Cs using t5hem
but that's about it: I can't thing of an A or B that tried them, but they
probably don't add much in RN number range (40,000 - 100,000). What does
seem to work is the Hoerner tip in conjunction with a swept LE on the
outermost wing panels. as well as the effects Schueman described, fairing
the upper surface so it meets the lower surface at an acute angle, raking
the tip at least 30 degrees and making it meet the TE at a sharp point
all helps to reduce the tip vortex, maybe moves it outboard a bit and
certainly stabilises its location, all of which seem to help thermalling
performance and certainly don't hinder the model's ability to self centre
in thermals.

I notice that the F5B and F5F boys are using similar wing planforms to
current FF models, i.e usinh Hoerner tips rather than winglets and are
using them at a very much higher RN, combined with interesting low drag
aerofoils that differ radically from anything I've seen on either F1ABC
models or our bigger toys. They obviously glide and thermal well and do
so at surprisingly high airspeeds.


> Gliders with span
> limits (for class or structural weight considerations) still generally
> have a vertical winglet at the tip rather than the Boeing-style flat
> raked tip (though a winglet and a span extension have similar effects on
> wingtip vortex and induced drag reduction for slightly different
> tradeoffs in bending moment).
>
That's interesting. Maybe, if the class rules impose a span limit its
better to accept the additional surface drag of the vertical tiplet
because of the extra area you get from the blunter tip?



--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

[[email protected]]

"Reality trumps theory" - Carlos Wu

If the polyhedral wing ships win a lot there's probably something to it.

[[email protected]]

Yes, but the feelling of the thermal is worse than with Discus1

[[email protected]]

I had 9 years of Discus B flying and 8 years in a Discus 2cT, which has the most sweepy- uppy wings of any current glider, and for me the latter had by far the better thermal feel and feedback.