Airfoil selection for an aircraft out there?

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

WARPilot

Well-Known Member
HBA Supporter
Joined
Apr 20, 2019
Messages
73
Your info and analysis is greatly appreciated. The 37A315 seems to offer the most desirable characteristics. I would like to still move to a 12% section at the tips and add the slight amount of twist(2 degrees). That may change if I can see what the 315 looks like at the tip.

The ULPower engine is 130hp and 170# plus misc parts such as baffles. This is the easiest way to build quickly with minimal shape changes. The airplane is known to be a good flyer but with an abrupt stall. The Riblett airfoil should help that somewhat. I loose some efficiency but gain some speed and better stall.
 

WINGITIS

Well-Known Member
Joined
Jun 24, 2020
Messages
634
Location
Wellington, New Zealand
NACA ran wind tunnel tests at lower Reynolds numbers than we see in TOWS. NACA TR 586 "Airfoil Section Characteristics as Affected by Variations of the Reynolds Number" contains test results with Reynolds numbers as low as 40,000. Polars for the 2412, 4412, 4415, and 23012 and a variety of other airfoils are shown.
Airfoil Section Characteristics as Affected by Variations of the Reynolds Number - NASA Technical Reports Server (NTRS)
THANK-YOU for posting the link to that paper Marc.

It is VERY GOOD in many ways! it even has one or two Reflexed Flying Wing airfoils covered in the data.

It could very well take up a few WORTHWHILE hours of everyones time...

Regards
Kevin
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
But if a wing is straight and tapered, the two most common ways of getting the root to stall first are:
- Washout (twist)
- Varying the airfoil thickness from root to tip
Two more ways:
  • Simplest way is to use one airfoil root to tip. With the chord tapered and the % thickness not tapered, they tend to behave fine.
  • Another way is aerodynamic washout, where the tip airfoil is same family, but just the next deeper camber. For instance, root at 63-215, tip goes to 63-415, root reaches stall before tip does, and stall in 63-415 is softer too.
Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
Your info and analysis is greatly appreciated. The 37A315 seems to offer the most desirable characteristics. I would like to still move to a 12% section at the tips and add the slight amount of twist(2 degrees). That may change if I can see what the 315 looks like at the tip.
What are you attempting to do by going thinner at the tips?
  • In the real world, you will have a hard time ever telling the difference in cruise drag between a 15% thick foil and the 12% thick foil;
  • The weight save on a composite wing that tapers its % thickness too is about imaginary as the spar is down to min gage on the web and the cap long before you get near the tip on either one - maybe 0.1 pounds - you can save a lot more than that with judicious design and careful adhesive application;
  • If it is just looks, the rounded style of wingtips used on the -190 will obscure the airfoil shape and thickness way beyond anybody's ability to do so.
If you are still convinced that there is pay value in going thin at the tip, it will take either: Geometric washout (2 degrees might not be enough) or a more highly cambered foil (37A412 or 37A612) to get the tip to play nice where the root runs at the edge of stall. With other foil families (NACA 63- and 64-) aerodynamic washout starts to require a lot more camber to get back to something resembling reasonable behavior.

Go with Riblett foils, and they all stall soft, then using one foil from root to tip works nice according to all the folks who have used them. That includes the Bearhawk designer Bob Barrows and our own late Bill Husa (Orion).

Billski
 

Lendo

Well-Known Member
Joined
Feb 6, 2013
Messages
904
Location
Brisbane
BigL, the GA 37A315 is a good selection, I believe Billski is using that in his design, I selected the GA 37A 415 for my design. Billski chose his for what ever reasons but I notice the Cm/C4 is lower (less pitching moment), I chose mine for the higher lift for lower landing speed. I believe Orien used the 35A 415 on one of his designs and an Aeronautical Engineer friend has chosen that for a single seat design.
I mention these people as their very experienced engineers, with the exception of myself, so just maybe they know something and why I take notice to what they have to say.
George
 

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,510
Location
Clatsop, Or
There are some things I'd like to see in a para-laminar airfoil.
Maximum height of the upper surface moved to the 33% chord position, for spar placement. Suggested thickness increase to 18% root, 13.5% tip with increasing camber matched to washout. A better representation of flap effects set in the up position. In terms of added drag increment, it's going to be very hard to measure it. 12% symmetric tail sections are common.
 

WARPilot

Well-Known Member
HBA Supporter
Joined
Apr 20, 2019
Messages
73
What are you attempting to do by going thinner at the tips?
  • In the real world, you will have a hard time ever telling the difference in cruise drag between a 15% thick foil and the 12% thick foil;
  • The weight save on a composite wing that tapers its % thickness too is about imaginary as the spar is down to min gage on the web and the cap long before you get near the tip on either one - maybe 0.1 pounds - you can save a lot more than that with judicious design and careful adhesive application;
  • If it is just looks, the rounded style of wingtips used on the -190 will obscure the airfoil shape and thickness way beyond anybody's ability to do so.
If you are still convinced that there is pay value in going thin at the tip, it will take either: Geometric washout (2 degrees might not be enough) or a more highly cambered foil (37A412 or 37A612) to get the tip to play nice where the root runs at the edge of stall. With other foil families (NACA 63- and 64-) aerodynamic washout starts to require a lot more camber to get back to something resembling reasonable behavior.

Go with Riblett foils, and they all stall soft, then using one foil from root to tip works nice according to all the folks who have used them. That includes the Bearhawk designer Bob Barrows and our own late Bill Husa (Orion).

Billski
Yes it is the overall look. I am unfamiliar with what the sections will look like if the same section continipues the entire span. I don’t want it to be really thick on the ends. It may not be…….I am just trying to visualize it not knowing what the change is as the chord shortens.
 

WARPilot

Well-Known Member
HBA Supporter
Joined
Apr 20, 2019
Messages
73
This has been very educational. Now can I inject an elliptical wing and ask what is difficult or different for getting good flight characteristics? I have heard there are different challenges. Would the same GA 37A315 work equally well?
 

Attachments

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
The -190 planform has a taper with a speed strake near the root. To give the right look, you would emulate those features. By the time Kurt Tank was leading the design team for the -190, the drawbacks of elliptical wings and elliptical tips on tapered wings were well known. Just smoothing over the end of the tapered wing had been found to make a better wing and that was done on the FW-190, P-51, F8F, and other contemporaries. Simple to tool and build in sheet metal too.

The elliptical plan form was thought to be perfect as lift and pressures have strong tendency to be elliptically distributed and give constant downwash. Drawbacks exist. The big round or elliptical wing tips tend to pull tip vortices inboard, and the wing then has induced drag like it has less span than it actually has while having all the skin friction of its full area. It is also more expensive to tool and build in metal or plywood - the skins must be compound curved. It turns out that constant taper with a simple tip is better aerodynamically.

There are other reasons to use elliptical wings. Packaging is big. Places to put magazines to feed the guns drove a lot of the design of the Spitfire and Thunderbolt. Fuel too. And a close look shows that the elliptical shape was not carried all the way to the tips in the Thunderbolt nor in later Spitfires either, but was cut off and edges smoothed over.

In composites and with 3D milled tools, you can put whatever shapes you want pretty easily. That does not mean you should do things that make for poorer performance.

I would emulate the look of the real thing in overall planform, shape of control surfaces, tips, etc, but put in one modern airfoil from root to tip, and let the rounded over tips camouflage my airfoil choices. The only way folks would easily see the foil choices is if you just use a square cut tip like the early Mooney's, which you surely would not do.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
I believe Billski is using that in his design
I selected 37 series for several reasons:
  • 35 series has a wider drag bucket than I will use;
  • 37 series has a drag bucket that I will be in during climb, throughout cruise, and during fast descents. Its laminar flow has some tolerance for accuracy errors if adequately smooth and fair;
  • 40 series has too narrow a drag bucket and it laminar flow is thought to be pretty sensitive to accuracy;
As for the rest of selection:
  • I selected the non-cusped aft section to keep aileron forces down, thus the A
  • My cruise Cl runs from 0.3 to 0.35, so the 3 in 315;
  • And 15% because that made for a light enough wing structure, and enough fuel volume;
    • Well, I thought it did. By the time I was done, I had fuel ahead of and behind the spar, and ran it well out the span. I will forever have to do two fill cycles to get a full fill. If I were doing it again, I would have made the wing 17-18% thick. My structural weight would have been lower and my fuel could have been carried all on one side of the spar.
Anyone who thinks that thick foils make for much drag should take a quick look at the Boomerang. 17% thick wings.

Billski
 

User27

Well-Known Member
Joined
Apr 5, 2015
Messages
132
Location
England
Sorry to be late to this thread ... With moulded construction it is quite a risk for a new design team to build moulds right away for a wing. Moulds are expensive, the tooling foam is costly as is the CNC router time and finishing. Might be better to build a plywood wing first to ensure you have good handling and performance characteristics, doesn't need to be aerobatic so stress calcs can be conservative to keep Turweston happy. You will make some mistakes if this is your first attempt. Look at fast construction techniques - foam or routed plywood ribs, strip carbon & foam spars, plywood skin perhaps - that might give a small weight penalty to get a proof of concept aircraft flying. Fisher Flying products produces some good videos that discuss light weight vs quick ribs for wood wings. Everyone wants to go with a carbon & cored wing and fuselage, they are light, but tooling and set up is expensive. If you make a mistake in the design - for example you are 5% heavier than expected and wing is actually 5% too small, fuselage is 5% too short, you may be completely stuffed! Make a proof of concept aircraft, and perhaps use RAeS E conditions to fly, and you may end up saving yourself a *lot* of money and some time.

I'm another fan of Riblett aerofoils, his book is still available from EAA and is worthwhile, just ignore the BS sections and also non-tapered tip sections, a rounded tip will hide a slightly thicker section. TOWS is also easily available.
 

WARPilot

Well-Known Member
HBA Supporter
Joined
Apr 20, 2019
Messages
73
What does “mod” mean in a wing foil section? For example….NACA 64-415 mod. I am assuming it means modified. How do I look at the shape? This is not for the WAR project. I have a Grumman and when I was looking at the incomplete guide to airfoils, I thought I would look up my Grumman’s.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,135
Location
Memphis, TN
It means someone changed the way they wanted to and started with xyz airfoil. I do know the Grumman airfoil mod is essentially the AA-1 airfoil but they kept the top airfoil profile but they have the top intersect with the bottom flat extended forward. Essentially they filled in the semi symmetrical bottom part of the leading edge of the AA1 airfoil. AA-5 same.
 

raymondbird

Well-Known Member
HBA Supporter
Joined
Nov 29, 2008
Messages
129
Location
Belleville, Ontario Canada
Going back to Chris's earlier post with performance specs for the WAR FW 190, the recommendation could change a little. Averaging out the top speed, climb rates indicating prop selection and factoring in the Verner Radial engine that may pose a problem with Cooling Drag, a safer route to go would be the (if you want) Riblett GA 37A-315.

Still believing that a weight of 924 lbs can be achieved (noting the aerobatic specs for G Loads requiring a certain level of structure to maintain strength) but airspeed may realistically be in the 160 mph area; a Lift Coefficient at that speed and weight would be 0.21 which fits the GA 37A-315 almost perfectly, or should we put it the other way round.

The maximum Coefficient of Lift (Cl) would be 1.49. Really just a small gain but the airfoil's Drag Bucket extends up to a (comparatively) higher Cl of 0.9 (from 0.83). Yet still reaches down to 0.03. That is usefully less than needed.

The no-Flap stall speed is around 56 mph.
But, is a pointy nose laminar wing going to look right on a ww2 replica? Except for the Mustang of course. I don't think so but maybe that is just me . . . ?
-replica 109 builder
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
Is foil chosen important to the looks? I do not know. I know that the cockpit is out of scale or is otherwise changed in these replica fighter, as is the engine, the planform, linkages/fasteners, prop, landing gear, wheels/tires/brakes, the list goes on and on.

I suggest that since you are making a fun to fly airplane, it will not be a scale model of the real thing for static display. Perhaps you make a list of everything that will not exactly scale down or be proportionally correct or might give better performance, then detail the errors and prioritize them. Maybe survey your customer base on their priorities too.

I suspect that few will prefer a slower airplane with awful stall behavior and a tendency to snap roll from a turn. Just because the original did that, does not mean you want to replicate that.

You can be faster with good stalls up into maneuvering flight with what I think will be a modest "edit" to bird's look compared to the many things that must be off to be practical.

I suspect that standoff appearance of this mini fighter will be achieved with fuselage and canopy form and proportions, canopy that looks right when open, wing and tail planform and proportions, ground stance, and the visuals. Make your demonstrator with authentic feeling paint scheme, details put on with an airbrush and shading, perhaps even some faux paint wear around the cockpit or smoke deposits airbrushed on around where the gun muzzles and exhaust pipes are on the real thing. All stuff that conjures the photos and films we have seen of these birds. Do that and I think you will grab folks' attention.

Make it fly slow and stall poorly and snap roll out of a steep turn, and the authentic wing design errors faithfully reproduced will not matter much.

Billski
 
Last edited:

raymondbird

Well-Known Member
HBA Supporter
Joined
Nov 29, 2008
Messages
129
Location
Belleville, Ontario Canada
Very good points as always Bill. It probably is just me that would notice that big, fat, round leading edge (naca 5 digit) or lack thereof. I used to be into scale RC though but really, isn't the 5 digit actually petty hard to beat aerodynamically? Needless to say judging by all the great flying machines that use it.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,584
Location
Saline Michigan
OK, raymondbird, you asked....

23012 through 23015 on a straight wing, like on the Van's series, have good stall behavior. It is a short span straight wing, and short span straight wings with the same airfoil from root to tip stall nicely. In an amateur built sheet metal wing, that foil in short span straight wing works OK.

You on the other hand are talking a composite wing, and so can get the accuracy and smoothness required for laminar flow and can get a significant drag reduction over the 230xx foils with NACA or Riblett laminar flow foils. You are also doing a tapered wing with a speed strake and less than optimal wing tips to emulate the FW-190 planform for that emotional grab of fighter replicas. Tapered plan form and tapered % thickness is well demonstrated to have poor stall progression and poor behavior. Folks are going to maneuver these birds. You really do need good stall behavior both for landing and for maneuvering. Using the 23015/23012 on a tapered wing is stacking the deck in favor of poor stall behavior. Washing out the wing can make for nasty stall/spin behavior from negative g. If you go ahead with it this way, be prepared for poor stalls, and then the whole scheme of fixes might be needed and hoping they are enough. Stall strips inboard, VG's or GlasStar style devices outboard, and so on. Those fixes may be way more obvious and distracting to the customer and the critics than just using a modern airfoil.

As to a modern airfoil being obvious, the tip of the -190 is like it was cut off square and then the sharp edges sanded round, somewhat disguising the airfoil. At the root, there is a speed strake and generous wing root fillets, and these will also nicely disguise whatever foil you choose, particularly if their shapes at the fuselage look like the original. I do suspect that when these features are replicated in your bird, the foil details will be pretty subtle indeed. In exchange for good behaviour without fixes stuck all over the wings and more speed, I think they are a good trade.

Perhaps building a 1/8 scale model of the bird both ways, hitting it with some camo paint, and look to see how different it really looks. This sort of exercise can help you also with handling other details that will have to change with scaling. Maybe there are some really accurate scale models of the real thing that you can build one straight up and then another modified to your flavor for overall comparisons. Letting other folks look at them is in order too. Or do them as solid models on the computer, complete with camo paint and root fillets and simple wingtips.

I have been stopping short on one other topic, but will now get to it. You might want to make sure that your bit of originality bias - "the original had a fat leading edge and this one doesn't" - has value in your potential customers before you commit to something. There will be other things that do not scale too well either, and they will also need to be either accepted or disguised. Your choices on each. Building one just for you is different from trying to sell something that may bite the customer.

Billski
 

AeroER

Well-Known Member
Joined
Oct 6, 2021
Messages
189
Sample single seat aircraft fuselage molded. Very similar in size to the WAR
Is there a finished, flying airplane using these parts?

The best approach to scaled WWII era piston engine fighters is to start with a cockpit sized for real pilots, then add the desired styling cues from the original around the pilot. You've already mentioned the radial engine candidate was too large to fit inside a WAR cowling.

The originals are not terribly large airplanes, and linear scaling much smaller than 80% to 85% results in an unusable cockpit volume.

I can barely get in an original BF 109, and I'll bet a FW190 is not much different. I haven't been in one, that might have been because I was afraid of getting stuck when I tried, I don't recall.

What is the plan for scaling the wheels and tires?

I'll make one more comment; selecting graphite simply because the eventual manufacturer sells graphite is the wrong reason. Doubly so if the plan includes selling enough kit parts to pay for the tooling cost. Counting on miracles of light weight and reduced drag is silly; make a realistic estimate of the weight by putting pencil to paper to make sketches and calculations. You'll get a surprise with a crude estimate before any sizing has been done.

I have worked on exactly one gr/epoxy part that was lighter than the aluminum part. That was the shear webs in the forward keels in Bird of Prey. They were lighter because the thickness squared effect on buckling over rode the differences in density and the parts had few or no penetrations except for fasteners.
 
Last edited:
Top