Logo
Builders' Plan Gallery  |  Hip Pocket Web Site  |  Contact Forum Admin  |  Contact Global Moderator
January 17, 2017, 10:05:50 AM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with email, password and session length
 
Home Help Search Login Register
Pages: [1]   Go Down
Print
Author Topic: symmetrical foil with simple flaps vs multi element flaps, Re from 10k to 100k  (Read 1558 times)
0 Members and 1 Guest are viewing this topic.
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« on: October 25, 2014, 04:37:02 AM »

Can anyone point me to some papers or data relevant to this subject? Only thing I've seen is an article by Francis Reynolds on a multi element setup for a wing sail on a model boat. I'm not sure I believe his Cl number and he says nothing about drag. See attached illustration.

The application I'm thinking about doesn't necessarily need quite as high a Cl, as chord can be increased as necessary, but a decent L/D would be nice.

Thanks

Attached files Thumbnail(s):
symmetrical foil with simple flaps vs multi element flaps, Re from 10k to 100k
Logged
USch
Gold Member
*****

Kudos: 11
Offline Offline

Italy Italy

Posts: 720




Ignore
« Reply #1 on: October 25, 2014, 05:44:56 AM »

A good starting point would be the classic "THEORY OF WING SECTIONS" by Abbot and Doenhoff, Dover Publications inc., 1949 (ISBN 486.60586-8). Most airfoils are tested without flap and with a 20%C flap, 60° down.
There must be some other NACA reports with more diversifyed flap deflection angles.

Urs
Logged

Fast up-Slow down
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #2 on: October 25, 2014, 02:06:45 PM »

Thanks. I've got that one. Unfortunately, the Reynolds numbers covered are much too high. I've also got Profili, which doesn't do multi element airfoils. NACA  report 93 gets closer, but not close enough unless I made a model too big to carry. Plus I don't think I would trust their tunnel to be sufficiently turbulence free.
Logged
ricardo
Silver Member
****

Kudos: 9
Offline Offline

Australia Australia

Posts: 188



Ignore
« Reply #3 on: October 25, 2014, 04:33:19 PM »

I think you are on virgin territory.

The advanced (?!!) versions of Xfoil do multi element foils.  Might be called something else.

Drela says Xfoil is good below Re=10k and above about 100k.  It's the 30k - 50k region, critically dependent on separation bubbles, where it is most iffy.

But Xfoil is still the best guess in this region.  Alas, that's where most of our models fly.

Try the Yahoo Xfoil group.  Mark Drela lurks there.
Logged

An engineer is someone who can do for 2 bob what any fool can do for a quid
TimWescott
Gold Member
*****

Kudos: 11
Offline Offline

United States United States

Posts: 830



Ignore
« Reply #4 on: October 25, 2014, 07:26:50 PM »

You might try Martin Hepperle's site, if it's still out there.  His airfoil thingie did multi-section foils (even to the extent of calculating biplane sections, treating both wings as one).

There's a fellow using flaps similar to that for CL stunt.  I'm not sure what his claimed advantage is, but he seems to be pretty adamant that it's a Good Thing.
Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #5 on: October 25, 2014, 10:54:13 PM »

I'm glad you mentioned that. Yesterday I was messing with it and getting some kind of bug that gave me implausible numbers. Today I was able to get it to behave some of the time, but the numbers seem fairly optimistic. I got slightly better numbers than Francis Reynolds was claiming. I also tried a more feasible two element airfoil, which was pretty good, but so was a NACA 009 with a simple flap. Not sure what to believe. In the limitations section, Mr. (Dr.??) Hepperle says that Javafoil is not good with separation and separation bubbles, so I don't know how much to trust any of this.

I tried the NACA 009 with flap in Xfoil (actually Profili) and got much less maximum lift and somewhat less L/D.

So I'm still wondering how much of this reflects reality.

I probably should go over to the Xfoil group. I'd forgotten about it. Thanks!
Logged
TimWescott
Gold Member
*****

Kudos: 11
Offline Offline

United States United States

Posts: 830



Ignore
« Reply #6 on: October 26, 2014, 12:30:16 AM »

I wonder how hard it is to build a model-sized wind tunnel, and keep it accurate?
Logged
ricardo
Silver Member
****

Kudos: 9
Offline Offline

Australia Australia

Posts: 188



Ignore
« Reply #7 on: October 26, 2014, 05:53:32 AM »

In the limitations section, Mr. (Dr.??) Hepperle says that Javafoil is not good with separation and separation bubbles, so I don't know how much to trust any of this.
Hepperle's stuff is the Eppler code updated a bit.  He explains this somewhere.

It only warns if separation is a big factor but doesn't really make a better guess as to what's happening.  Martin Simons's later editions have some details.  I don't think any version of Eppler can be trusted below Re 100k

I think Xfoil is unique in giving quite good guesses at model Re.

Martin Simons's book has the caveats about doing your own Wind Tunnel for low Re.
Logged

An engineer is someone who can do for 2 bob what any fool can do for a quid
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #8 on: October 26, 2014, 06:14:52 AM »

I wonder how hard it is to build a model-sized wind tunnel, and keep it accurate?
I'm pretty sure it's fairly difficult. You've got to keep the turbulence levels very low. Might be easier to make some kind of flying model, but that presents it's own problems.
Logged
Ara Dedekian
Bronze Member
***

Kudos: 5
Offline Offline

United States United States

Posts: 85



Ignore
« Reply #9 on: October 29, 2014, 10:49:45 AM »


Reply #4

There's a fellow using flaps similar to that for CL stunt.  I'm not sure what his claimed advantage is, but he seems to be pretty adamant that it's a Good Thing.
Attached files Thumbnail(s):
Re: symmetrical foil with simple flaps vs multi element flaps, Re from 10k to 100k
Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #10 on: October 29, 2014, 12:50:35 PM »


Reply #4

There's a fellow using flaps similar to that for CL stunt.  I'm not sure what his claimed advantage is, but he seems to be pretty adamant that it's a Good Thing.

Hi Ara,
It ought to work very well. Probably his square loops have sharper corners than anyone's. However, if his model is going 5O mph, with a 9 inch chord, the Reynolds number will be around 350k.
Logged
Terry Fitzpatrick
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 34



Ignore
« Reply #11 on: April 29, 2015, 06:40:51 AM »

What is the application for this Lincoln? I have an idea for "hard" model yacht sails with leading and trailing edge flaps to try in the future. The idea is based on low Re number wind tunnel data on circular arc aerofoils.

regards
Terry Fitzpatrick
Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #12 on: April 29, 2015, 07:25:15 AM »

The application I have in mind may be almost the same as yours, though I'm not sure a circular arc is the best way to go.

An additional feature I'd like to try might be a vane, or a vane, pot and servo to hold the "sail" at a constant angle of attack rather then waiting for the skipper to notice the wind shift and respond. I don't know about anyone else, but it seems to me that, on those rare occasions when I sail a model, it's out of trim a very large fraction of the time.

Might be just the thing for a footy. A hard foil might have much less drag when tacking, which might enable lighter boats that could still tack effectively?
Logged
ricardo
Silver Member
****

Kudos: 9
Offline Offline

Australia Australia

Posts: 188



Ignore
« Reply #13 on: April 30, 2015, 02:38:38 AM »

A hard foil might have much less drag when tacking, which might enable lighter boats that could still tack effectively?
Modern Sailboard (Windsurfer) sails are already like this.

Fully battened with shaped batten keepers that sit on the mast in a large luff mast pocket so they are like Jedelski foils but with even nicer curves & shapes.

Very sophisticated.  Low drag is important cos that make the sails MUCH easier to handle.  Washout that increases with windspeed etc.  Try some of the Sailboard forums.
Logged

An engineer is someone who can do for 2 bob what any fool can do for a quid
dosco
Gold Member
*****

Kudos: 5
Offline Offline

United States United States

Posts: 786



Ignore
« Reply #14 on: April 30, 2015, 07:58:33 AM »

You've seen the North Sails 3-dimensional "molded" sails?

-Dave

Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #15 on: May 01, 2015, 11:39:11 PM »

Never heard of them. I've seen some of the more recent sailboard sails, and they certainly are impressive.

One of the advantages of totally rigid sails is that it's relatively easy to get the shape right. Also, I suspect they're easier to design, unless you want to get twist without an extra channel.
Logged
Terry Fitzpatrick
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 34



Ignore
« Reply #16 on: May 02, 2015, 02:03:25 AM »

Attached is a proposal drawing for a low Reynolds number aerofoil with equal performance in both directions i.e. for  yacht applications.The basis for the aerofoil is the good performance of thin cambered sections at low Re. The reason for this according to  F.W. Schmitz (Model Aircraft Aerodynamics,by Martin Simons) is  small leading edge radius (0.7 % chord used in my drawing) & small upper surface curvature . I have used the circular arc as the basis because it gives the flattest upper surface for a given amount of camber . I have wind tunnel data at Re = 320,000 (I thought it was lower than this !!!). I think the idea could be adapted to other thin cambered aerofoils with low Re. credibility as well.  My intention is to build a sail of this type for the RM 50/800 class boat (I have one), but its been on the "to do" list for several years like many other fun projects I want to do. The camber could be increased by reducing the 320.5 mm radius to a smaller number, or by deflecting the flaps further so they are no longer tangents to the main surface. I was thinking that the tangent approach would give the best L/D for good performance into the wind.

There are probably good reasons for making the flap angles adjustable when sailing. My original idea was to have the flap deflection angle set at a single value and the flaps free to flop over as required by the boats heading relative to the wind.
 
I will see if I find some supporting wind tunnel data (including the circular arc data) and send that as a follow up.
 
regards Terry Fitzpatrick
Attached files Thumbnail(s):
Re: symmetrical foil with simple flaps vs multi element flaps, Re from 10k to 100k
Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #17 on: May 02, 2015, 03:29:11 AM »

My intuition says that it might work pretty well on a bigger model like that. I was thinking about smaller boats, so my idea, after playing around with Xfoil for a while, was an airfoil that was nearly a flat plate, with 15 percent flaps on both ends, with up to 15 degrees deflection. In order to use just one channel to control the sail, I was going to have flap deflection move along with the vane which controls angle of attack.

Reducing camber might be useful when the wind gets fresher and you have to operate at a lower angle of attack to prevent excessive healing. (Kind of like easing the sheet on a normal sailboat.) I suspect that one could tack closer to the wind with a rigid sail, also that it would have less drag while tacking than a flapping conventional sail.

Another scheme I have seen in a couple of places is to make the sail symmetrical, so that you turn it around to keep the camber going the right way. But there didn't seem to be any way to reduce camber. Simpler that way, but makes the sail more of a vulnerability in strong winds.
Logged
Terry Fitzpatrick
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 34



Ignore
« Reply #18 on: May 02, 2015, 06:12:58 AM »

My thoughts on coping with strong winds would be to fit a smaller sail (storm rig) or have the top 300 mm of the hard sail removable. What size boat do you have in mind, and what sail chords? Another advantage of the hard sail is the shape can be better than the usual triangle with large amounts of twist. It should come much closer to an elliptic lift distribution.

regards Terry Fitzpatrick
Logged
lincoln
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,561

Topic starter


Ignore
« Reply #19 on: May 02, 2015, 06:23:23 PM »

I was mostly thinking about a footie, but I've also thought about a larger boat to use in an event that's pretty much anything goes. I think footies should be much lighter, and maybe something like this can make tacking easier, so that a lighter boat can avoid being caught "in irons".  Maybe a "skimming dish" design, proportioned much like those singlehanded racing boats, might be good.

I don't remember the sail chords any more.

I also had some ideas for a one meter, very narrow, light displacement schooner. Differential sheeting would help it tack.
Logged
ricardo
Silver Member
****

Kudos: 9
Offline Offline

Australia Australia

Posts: 188



Ignore
« Reply #20 on: May 03, 2015, 07:18:23 AM »

Reducing camber might be useful when the wind gets fresher and you have to operate at a lower angle of attack to prevent excessive healing. (Kind of like easing the sheet on a normal sailboat.) I suspect that one could tack closer to the wind with a rigid sail, also that it would have less drag while tacking than a flapping conventional sail.

Another scheme I have seen in a couple of places is to make the sail symmetrical, so that you turn it around to keep the camber going the right way. But there didn't seem to be any way to reduce camber. Simpler that way, but makes the sail more of a vulnerability in strong winds.
You certainly sail closer to the wind with a fully battened sail.  The camber flips over when you tack.

It's also possible to adjust camber using the outhaul & downhaul but not when sailing (a sailboard).

Go and have a look at some sailboard sails in the flesh and in use.
Logged

An engineer is someone who can do for 2 bob what any fool can do for a quid
Terry Fitzpatrick
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 34



Ignore
« Reply #21 on: May 03, 2015, 07:52:06 AM »

Whats a footie? A foot as in 12 inches long hull? I have seen some very old  racing model yachts from 1920 or 1930 that were raced here in Sydney. They had huge deep bulb keels, & massive sail area, a very short beamy flat hull. Is that a footie?

regards Terry Fitzpatrick
Logged
Terry Fitzpatrick
Bronze Member
***

Kudos: 0
Offline Offline

Australia Australia

Posts: 34



Ignore
« Reply #22 on: May 03, 2015, 08:16:28 AM »

I googled "footie " and come up with the English "footy" at a model yacht website. I thought the 36R as slightly smaller than a sensible size! but no..... there is the footy! I think a skimming dish would be a great hull shape for these  little boats. The old boats I have seen had bows that were very blunt, perhaps hemispherical and with there huge sail area could probably plane.

regards Terry Fitzpatrick
Logged
basiliscus
Nickel Member
*

Kudos: 0
Offline Offline

United States United States

Posts: 1



Ignore
« Reply #23 on: December 24, 2016, 06:26:02 PM »

Can anyone point me to some papers or data relevant to this subject? Only thing I've seen is an article by Francis Reynolds on a multi element setup for a wing sail on a model boat. I'm not sure I believe his Cl number and he says nothing about drag. See attached illustration.

The application I'm thinking about doesn't necessarily need quite as high a Cl, as chord can be increased as necessary, but a decent L/D would be nice.

Here are some data on a single-slotted wingsail section similar to what you've shown: http://www.tspeer.com/RigidRigs/50flap/S901fa20.htm
These results were done with the MCARFA program, which is very dated but the best I had at the time.

You can use Javafoil, http://www.mh-aerotools.de/airfoils/javafoil.htm, to design multi-element sections.  However, Javafoil has a couple of drawbacks for working at model Reynolds numbers.  The first is it cannot handle laminar separation bubbles.  This is a serious deficiency because it's not possible to get natural transition to turbulent flow without laminar separation when operating at Reynolds numbers below about 250,000.

The second drawback is Javafoil cannot predict separation that occurs in the middle of the flowfield.  Separation for a single element section will always occur on the airfoil surface, typically because of the increase in pressure toward the trailing edge, or as a result of the sudden increase in pressure behind a leading edge suction peak.  But a multielement section can have a situation where the flow gets backed up in the wake of one element as it passes over a trailing element.  For example, the wake behind the thick main element of the airfoil shown can encounter increased pressure as it flows over the flaps, and the flow in the wake can back up leading to the outer flow not following the shape of the inner airflow layers.  This results in the flow through the slot(s) being attached to the flap surface like a wall jet, but the airfoil is still stalled.  I've seen this stall mechanism for well-tuned full scale wingsails, but I don't know if it would be the stall mechanism for model wingsails.

The MSES program can handle laminar separation bubbles and separation in wakes, but unfortunately it is a professional code that sells for a professional price.

I haven't designed any wingsail sections for model Reynolds numbers, but the principles would be the same as for full scale wingsails.  It could be interesting to see what MSES says about wingsails under model conditions.  I suspect you'd want to make the flap a bigger proportion of the chord because you need more distance for the pressure recovery as the Reynolds number goes down.  You wouldn't need as much thickness for structural strength and stiffness as for a full-scale wingsail, so you could stand to scale down the leading element.  I'm not sure a double slotted flap would work all that well because the Reynolds number of the flap elements would be so small.

I think I'd be inclined to start with a single-slotted section with a main element that was 35% - 40% of the total chord, and thin, almost flat plate section for the flap.
Logged
Pages: [1]   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!