Discussion Thread: The design of a tailless flying wing

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,882
Location
Marion, Ohio
I see one problem with the spreadsheet.

If you try to make the wing larger it won't display all of it. Limited to +/-3M span.
 

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
3,306
Location
Grand Junction, Colorado
Well the tech report, NASA/TP—2016–219072, is a year past the initially anticipated release date but it's finally out.
[EDIT]Oops... wasn't supposed to be public for another month [/EDIT].


A fin on the wing at the inboard end of the elevon may not be a bad idea either, whether it's a floating tip or a trailing edge surface. With a BSLD the tip vorticity is spread out and indistinct (like a hurricane when it's breaking up) but you can see that the core of the vortex is at about 70% of b/2. Winglets harvest energy from the vortex so on this type of wing they belong there not at the tips. That vortex though might have something to do with the favorable roll/yaw coupling so I'm not sure how absorbing [some of] it with a fin would affect that couple. But then you've got fins for yaw damping and control so... so what. Speaking of roll/yaw coupling a guy named Edward Udens did some testing with elevon shape and compiled a table of yawing moments that is really important here.

[video=youtube;5W1pmqLHYm4]https://www.youtube.com/watch?v=5W1pmqLHYm4[/video]
 
Last edited by a moderator:

choppergirl

Banned
Joined
Jan 30, 2015
Messages
1,682
Location
air-war.org
Juan de la Cierva invented nothing, nothing I say! Mother Nature already invented your tailless flying machine... and she did it... completely blind.

 
Last edited by a moderator:

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
3,306
Location
Grand Junction, Colorado
I apologize to those of you who didn't get that paper before I removed the link, I'll post the official link when NASA releases it. Here's an older paper that's mostly slides from Al's presentations. Also a couple attachments showing the nonlinear lift distributions he's been working with. Sorry the axes aren't labeled, X is span and Y is gamma (non-dimensional local load). Am I nuts or does it really look like they're approaching a Gaussian normal distribution?
 

Attachments

  • Prandtl_Horten_Jones.jpg
    Prandtl_Horten_Jones.jpg
    25.9 KB · Views: 77
  • P-H-J.jpg
    P-H-J.jpg
    49.6 KB · Views: 80

RPM314

Well-Known Member
Joined
Mar 27, 2015
Messages
735
Location
NY, USA
To quote the mad hatter on the subject of being crazy, "all the best people are".

And no I think this is mathematically distinct from the Gaussian distribution. The Gaussian asymptotically approaches zero as x goes to +/- infinity, but ESLD and BSLD are just different powers of ellipses, I think. Visual similarity is there, but doesn't mean much.
 

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,775
Location
Bellingham, Wa
I don't think the question of distribution is really as critical as we tend to obsess about.

We know we'll end up significantly longer in span with fairly pointy tips.

Big question is how do deal with the controls, decoupled or mixed.

Are we more concerned with drag or aerobatics?

Are we using straight, tapered or custom flaps. Do they make for slots, or simple pivot?

Does this add drag bucket range or for min landing primarily.

LuPi
 

StarJar

Well-Known Member
Joined
Sep 6, 2011
Messages
1,723
Location
El Centro, California, USA
It seems to indicate proverse yaw, but the tips are twisted so much, that I think some L/D ratio is lost.
The proverse yaw seems to come from the lifting elevon ducking out of the airstream.
Seems like why have it so much in the airstream to begin with?
I'm a little disappointed with the whole thing unless someone can show me where I might be missing something.
I did just breeze through it once, in all fairness.
 

Norman

Well-Known Member
Joined
Nov 28, 2003
Messages
3,306
Location
Grand Junction, Colorado
It's about induced drag not profile drag and is illustrated in figure 2 of NASA/TP—2016–219072. Here's a slightly different perspective. The one with the white background is mine and the one with the black background is a more technically accurate version provided to me by Marko Stamenovic.
 

Attachments

  • tilted-vectors-2.jpg
    tilted-vectors-2.jpg
    18.6 KB · Views: 90
  • tilted_vectors.jpg
    tilted_vectors.jpg
    26.4 KB · Views: 123

StarJar

Well-Known Member
Joined
Sep 6, 2011
Messages
1,723
Location
El Centro, California, USA
It's about induced drag not profile drag and is illustrated in figure 2 of NASA/TP—2016–219072. Here's a slightly different perspective. The one with the white background is mine and the one with the black background is a more technically accurate version provided to me by Marko Stamenovic.
OK thanks, but what about overall drag.
The reason I ask, is because when I tried to simulate it on Xflr5 using VLM analysis, twisting the wing more reduced the overall L/D ratio, (although the low speed elevon control was quite better.)
The possible flaw in my simulation was that I used a constant twist, rather than having the inboard section at a near untwisted incidence.
So I suppose that could of messed up the results.
But I thought that you and some others were downplaying the benefits of Bowers' proposal, so I'm in sort of a quandry.
 

RPM314

Well-Known Member
Joined
Mar 27, 2015
Messages
735
Location
NY, USA
Type ten elevons

Hi, all. I've been designing my next RC plane with the help of XFLR5, and I confirmed a little pet theory of mine.

We're all familiar with the application of the type ten elevon profile as applied to BSLD wings, but I have wondered for a while if they could help with some of the problems of ESLD craft, like jumps in the lift distribution as trim changes and high loading of the tips at high alpha. And indeed, it seems they do.
Control.jpg Contstant_Chord.jpg Full_Taper.jpg
Left to right, a wing with no elevons, a wing with constant chord elevons at trim speed, and wing with fully tapered elevons extending 3/4 of the way inboard at trim speed. All at a CL of about 0.5, no twist, and viewed from behind.

All display a roughly elliptical lift distribution, but this can be deceiving. The first two have a downwash pattern that grows greatly towards the tips, but we know that an ESLD done "properly" is the result of a constant downwash. The last one, with fully tapered elevons like Uden's Type 10, has a downwash that is much more constant. Once it is twisted a bit at the tips, I can get it very close indeed to the "ideal".
It also shows a slight CLMAX and CL/CD advantage over the constant chord elevon type. That, plus the above CDi advantage without substantial twist, plus the stall starts at 1/3 of the half-span, plus no sharp jumps in the spanload.... Seems like a very good deal.

So I'm getting a little trepidatious. It's been firmly hammered into my head that there are always trade-offs in this business, and I can't really spot any here. Can anyone think of any? Maybe there's gonna be some bad adverse yaw, since there's so much elevon area at the tips? Penny for your thoughts.
 

StarJar

Well-Known Member
Joined
Sep 6, 2011
Messages
1,723
Location
El Centro, California, USA
I also found much improvement by using them, albeit on X-plane. But my x-plane numbers have never varied much from xflr5.
Even on wings with only16° sweep (20° at LE) those elevons made a better wing, lower stall, more control.
I started hinging mine at 50% c at the tip, tapering to 0% or 5%. Xplane only goes to 50% c. (My wing was a .6 taper and 16° sweep.) Seemed to work great.
 
Last edited:
Top