HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Dec 09, 2012, 09:01 PM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
Discussion
Computer generated shapes

I've been playing with Blender, a 3D rendering program with some capability in CAD. I wanted a wing with curved dihedral and I got that. As I started playing, the thought occurred that some shapes seen on RC planes arent picked for their aerodynamic quality, but come to be more because they are done by computer.

The way I made my wing model was to extrude the basic airfoil curve by following 2 curves, one for planform and the other for taper. As I fiddled with these curves, I saw shapes I'd seen on other models, mostly sailplanes and some ARF park flyers.

Know what I mean, Vern?
HELModels is online now Find More Posts by HELModels
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 09, 2012, 10:20 PM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
I should give an example I guess. I'm not sure this shape is aerodynamically superior or inferior, but it is easy to generate with a computer. The pointed tip and the recurve on the traling edge seem characteristic of a computer. This is the shape I've seen on lots of foam parkflyers and some sailplanes.
HELModels is online now Find More Posts by HELModels
Reply With Quote
Old Dec 10, 2012, 12:10 AM
B for Bruce
BMatthews's Avatar
The 'Wack, BC, Canada
Joined Oct 2002
11,231 Posts
The "shark fin" like wing tip planform is done for aerodynamic reasons and not simply because it looks good or is easy to draw in CAD or whatever.

The shape is done to death these days in model designs where low drag performance matters because this shape tends to produce the smallest tip vortex shape. And a small vortex off the tip means less drag.

The fact that it's easy to generate in something like Blender or in CAD is just a side bonus.
BMatthews is online now Find More Posts by BMatthews
Reply With Quote
Old Dec 10, 2012, 12:46 AM
The Great Filter
kcaldwel's Avatar
Joined Jan 2007
3,534 Posts
Quote:
Originally Posted by BMatthews View Post
The "shark fin" like wing tip planform is done for aerodynamic reasons and not simply because it looks good or is easy to draw in CAD or whatever.
Any references for that Bruce? I can't find any in English at least. There is Van Dam's crescent shaped wing stuff which seems to be mostly an artifact of the analysis method he used, and the Schuemann straight trailing edge theory, but I haven't seen much to justify the shape that is used on so many sailplanes....

Kevin
kcaldwel is offline Find More Posts by kcaldwel
Reply With Quote
Old Dec 10, 2012, 04:22 AM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
Artifact, that's the word I was looking for.

Edit: What the... how did that get in there.

The combined curves which controlled the planform in the example pic reminds me of the path the ropes of lift actually follow on a wing - the function(pun intended)determines the form.
HELModels is online now Find More Posts by HELModels
Last edited by HELModels; Dec 10, 2012 at 01:59 PM.
Reply With Quote
Old Dec 10, 2012, 01:26 PM
Ascended Master
Sparky Paul's Avatar
Palmdale, CA
Joined Oct 2000
13,371 Posts
I did some tip vortex tests on a ZAGI a few years back... watched the erosion in length of cassette tapes on the wing tips, with and without tip plates.
The tip plates showed very little erosion after several flights, while the plain tip lost most of each tape during a flight.
Sparky Paul is offline Find More Posts by Sparky Paul
RCG Plus Member
Old Dec 10, 2012, 01:33 PM
B for Bruce
BMatthews's Avatar
The 'Wack, BC, Canada
Joined Oct 2002
11,231 Posts
Kevin, the first semi serious write up I saw was back in the 60's or early 70's in Model Airplane News. They did a set of wing tips and ran them in a wind or water tunnel with smoke or dye (sorry, don't recall which it was). Square ended, reverse cut and crescent or shart fin were the three tips tested. The cresent shape had by far the smallest and least turbulent flow.

I know you like numbers and math to back up this stuff but this particular report simply didn't have any, just the pictures.

Some years later, but still back quite a ways, I read through a Schumman related report in one of the model magazines where the crescent shape showed up again as an option as an aid in avoiding the formation of separation bubbles and some of the hysterisis in getting them to collapse and re-attach the airflow. Along with the swept back tips that we've come to call "Schumann" tips the crescent shapes were also considered. In that one the crescent shaped wing was again identified as one which would aid in reducing the tip vortex size and energy while being a shape which would be harder for the air to roll around and encourage span wise flow inward along the top of the wing. Again no hard numbers or actual wind tunnel testing.
BMatthews is online now Find More Posts by BMatthews
Reply With Quote
Old Dec 10, 2012, 02:15 PM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
Any idea if there is an advantage to a curved dihedral rather than simple or polyhedral?

The same bezier curve which controls the planform in x and y, when tweaked in z will make a curved dihedral. It's pretty, hard to build for real and easy to generate on a computer.
HELModels is online now Find More Posts by HELModels
Reply With Quote
Old Dec 10, 2012, 03:11 PM
The Great Filter
kcaldwel's Avatar
Joined Jan 2007
3,534 Posts
As with most aero, the obvious things don't necessarily produce the performance improvements you might expect. Tip plates do reduce the tip vortices, and the induced drag, but the parasitic drag of the additional wetted area and the intersections is far higher even at best L/D.

The Schuemann planform is just a straight trailing edge, with all the taper on the leading edge, not those aft swept wing tips that seem to be the subject of the thread:

http://www.spieltek.com/SunbirdSoari...ormArticle.pdf

I have the van Dam papers, which are for a crescent shaped wing planform. He used an elliptical chord distribution with a swept 1/4c line, which seems different than is typically used on models as mentioned here. Maybe the planforms used on RC sailplanes are just an artifact of the CAD design. This is one of the van Dam papers:

http://ntrs.nasa.gov/archive/nasa/ca...1989014047.pdf

Neither of those planforms takes into account the increased drag from the low Re tips at model scales.

The effects of planform shape on overall drag are very subtle, and not easily measured or modelled. I am very suspicious of most results from wind tunnel testing or CFD analysis. Most of the optimistic results have proven to be incorrect on further testing or analysis. I have some papers that indicate the optimum planar shape my be similar to the attached sketch. If the planform can be 3D, then there are a lot more options.

Kevin
kcaldwel is offline Find More Posts by kcaldwel
Reply With Quote
Old Dec 10, 2012, 07:40 PM
The Great Filter
kcaldwel's Avatar
Joined Jan 2007
3,534 Posts
Curved dihedral might be good, if done properly:

http://ntrs.nasa.gov/archive/nasa/ca...1964006060.pdf

Hyper-elliptic anhedral is better!

citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.117.7340&rep=rep1&type=pdf

Just as an example of what we are playing with on planform shapes, compare a rectangular wing with a couple of degrees of washout with the usual ideal elliptical span loading distribution wing. The difference is best L/D induced drag is less than 3%. At best L/D, induced drag and parasitic drag are equal. Reducing induced drag 3% means you have to reduce the parasitic drag by slowing down to get to the new best L/D speed. This increases the induced drag, so the overall aircraft drag might only fall by 2% at best L/D.

If we have a sailplane with a 25/1 best L/D, reducing the induced drag 3% might increase the L/D 2%, to 25.5/1. Over a 1km glide at best L/D, the rectangular wing sailplane would lose 40m (131.2'), and our elliptical distribution sailplane would lose 39.2m (128.6'). That is a difference of 0.8m (2.6'). While it is nice to have the improvement, it is just not visible in the real world of varying up and down air. It is even difficult to measure in good wind tunnels. And that is not considering the low Re effect on the smaller wing tip chords.

Increasing the wing span by a few percent should give a bigger improvement - the best tip design is to put them farther apart.

Of course there are other advantages to tapered planforms such as roll rate, structural, and aesthetic.

Kevin
kcaldwel is offline Find More Posts by kcaldwel
Last edited by kcaldwel; Dec 10, 2012 at 08:57 PM.
Reply With Quote
Old Dec 10, 2012, 10:13 PM
Grad student in aeronautics
United States, GA, Atlanta
Joined Oct 2010
433 Posts
Quote:
Originally Posted by Sparky Paul View Post
I did some tip vortex tests on a ZAGI a few years back... watched the erosion in length of cassette tapes on the wing tips, with and without tip plates.
The tip plates showed very little erosion after several flights, while the plain tip lost most of each tape during a flight.
Did you do a test with the tape at the tip of the winglet? I would like to see how that compared to the planar wing tip and the wing-winglet junction.
DPATE is offline Find More Posts by DPATE
Reply With Quote
Old Dec 10, 2012, 10:27 PM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
Quote:
Originally Posted by kcaldwel View Post
Curved dihedral might be good, if done properly:

http://ntrs.nasa.gov/archive/nasa/ca...1964006060.pdf

Hyper-elliptic anhedral is better!

citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.117.7340&rep=rep1&type=pdf

Just as an example of what we are playing with on planform shapes, compare a rectangular wing with a couple of degrees of washout with the usual ideal elliptical span loading distribution wing. The difference is best L/D induced drag is less than 3%. At best L/D, induced drag and parasitic drag are equal. Reducing induced drag 3% means you have to reduce the parasitic drag by slowing down to get to the new best L/D speed. This increases the induced drag, so the overall aircraft drag might only fall by 2% at best L/D.

If we have a sailplane with a 25/1 best L/D, reducing the induced drag 3% might increase the L/D 2%, to 25.5/1. Over a 1km glide at best L/D, the rectangular wing sailplane would lose 40m (131.2'), and our elliptical distribution sailplane would lose 39.2m (128.6'). That is a difference of 0.8m (2.6'). While it is nice to have the improvement, it is just not visible in the real world of varying up and down air. It is even difficult to measure in good wind tunnels. And that is not considering the low Re effect on the smaller wing tip chords.

Increasing the wing span by a few percent should give a bigger improvement - the best tip design is to put them farther apart.

Of course there are other advantages to tapered planforms such as roll rate, structural, and aesthetic.

Kevin
Thank you, that's a ton of information. Further apart is best, that can be built.
HELModels is online now Find More Posts by HELModels
Reply With Quote
Old Dec 10, 2012, 10:46 PM
Ascended Master
Sparky Paul's Avatar
Palmdale, CA
Joined Oct 2000
13,371 Posts
Quote:
Originally Posted by DPATE View Post
Did you do a test with the tape at the tip of the winglet? I would like to see how that compared to the planar wing tip and the wing-winglet junction.
.
I found a picture, but no information on the erosion difference. The tape is straighter than the tape on the other tip at the center of the plate.
Sparky Paul is offline Find More Posts by Sparky Paul
RCG Plus Member
Old Dec 11, 2012, 01:33 AM
Registered User
HELModels's Avatar
Joined Aug 2003
2,629 Posts
Here's the shape I settled on, it's only the last third of the wing. The mid span will be straight. It matches the tail shape I have so aesthetics win. One day it will get 3D printed and I'll use it to make a mold. The tradeoff is expense for ease and accuracy of building.
HELModels is online now Find More Posts by HELModels
Reply With Quote
Old Dec 11, 2012, 02:52 PM
B for Bruce
BMatthews's Avatar
The 'Wack, BC, Canada
Joined Oct 2002
11,231 Posts
I suspect that the big advantage to curved dihedral is that you lose the drag associated with the air traveling over the usual sharp breaks and forming turbulence.
BMatthews is online now Find More Posts by BMatthews
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Outstanding computer generated artwork Bob Bailey Scratchbuilt Indoor and Micro Models 10 May 20, 2006 07:48 AM
Gallery Computer generated R/C Model drawings Brettos Australia 38 Apr 22, 2006 08:22 AM