Avid STOL airfoil -some calculations

94 posts in this topic

Posted (edited)

Looking at my original post, it occurred to me that it's highly likely that in the cruise case the lower-surface flow, if it separates at all, re-attaches on its own as a turbulent layer, so it would make sense that nothing would help there.

Edited by Turbo

Share this post


Link to post
Share on other sites

Posted

I know what I need, you do not.

Share this post


Link to post
Share on other sites

Posted

Nick,

It's coming into focus for me, though.  More legroom, more speed, bigger gas tank(s).   My damn legs are too long!

Share this post


Link to post
Share on other sites

Posted

So, Manu, how is the new trim tab working out?

i don t know yet ...

Share this post


Link to post
Share on other sites

Posted

I was curious about the Avid STOL airfoil, so made some calculations using an inviscid panel method code.  The results are interesting.

 

We aerodynamicists tend to do everything we can to remove scale, or physical size from consideration.  For example, we divide the wing’s lift by everything it’s proportional to: wing area and dynamic pressure, to get a lift coefficient.  The dynamic pressure is just the difference between total pressure, that measured with a forward-facing Pitot tube, and freestream static pressure Pfs, that of the air at that altitude away from the influence of the airplane.  It turns out that at our speeds this difference is ½rVfs2, where r (Greek r) is the air density.  What’s great about total pressure is that it’s constant almost everywhere, except in regions of flow separation or where viscous effects are important, like right next to the surface, in the boundary layer.  Bernoulli would say that this means that in locales where local velocity is high, pressure will be low, to preserve the total pressure: Pt=Ps +½rV2.

 

 We talk about the pressure distribution around a body in terms of a pressure coefficient, Cp, which is just the difference between local pressure on the body’s surface and freestream pressure divided by the dynamic pressure

 

(Cp = (p-pfs)/½rVfs2).  When local airspeed, say, over the wing, goes up, pressure coefficient goes down.  That partial vacuum on the upper surface is a lot of what keeps us afloat so-to-speak.  But y’all already knew that!

 

So just for fun, I looked at not only the STOL airfoil shape, but a version for which a straight line defines the lower surface.  This is what Manu (Efil01) has on his Avid.  It has more camber than the speedwing airfoil, since the upper surface is more curved.  Here’s a plot of the airfoil shapes.

 

 

Avid_STOL_Airfoil.thumb.JPG.56723783f601

 

The first thing I looked at was the pressure distribution at takeoff, assuming the landing gear arrangement allows a maximum angle of attack of no more than 8 degrees, relative to the max length line (here the x-axis).  Taller main gear or a smaller tailwheel would help here.  By the way, the plot seems to chop off the trailing edge of the airfoil.  That’s just an artifact of what was plotted; the inviscid panel code used for the computation plotted the “control points” which are midway between the defining points.  I’m just being lazy here, so I beg your forebearance. 

 

So here’s what the Cp distribution looks like at takeoff:

Avid_STOL_takeoff.thumb.JPG.647933df1732

Per aerodynamics custom the Cp is plotted upside down, with negative values going up the ordinate, not down.  The blue is the original airfoil, the red is the flat-bottomed version.   At 8 deg AoA, the flat-bottomed airfoil delivers 7% less lift than the original version.  As you can see, the lower surface has positive pressure on it in both cases.  The upper surface features an overspeed right at the leading edge, perhaps the consequence of having too large an angular range for the round shape of the front spar tube, with a rather abrupt change in the surface curvature.  Note also that the upper surface, with its overall overspeed, contributes about 2.5  to 2.8 times as much lift (area under the curve) as the lower surface.

 

Still, the slowing down of the airflow on the upper surface is relatively gradual, which is a good thing (Thank you Dean!).  Note also that the lower surface flow is accelerating as it goes back towards the trailing edge.  However, the effect of the under-surface camber is about a 40% increase in nose-down pitching moment for the STOL section vs the modified shape.  All that camber is like flying around with the flaps down.  Yes, it gets us off of the runway faster, since with taildragger gear we’re angle-of-attack limited on the ground, and yes, it delivers glorious in-flight visibility over-the-nose.  But this is one of the great things about the Junkers-style flaperons:  The camber can effectively be increased for takeoff & landing, but that dang nose-down pitching moment can be banished by pulling flaps back up for cruise.  Having the wing and tail not fighting each other lowers the induced drags of both and allows us to cruise more efficiently, and faster

 

O.K. now, so what do the Cp plots look like in cruise?  I ran both airfoils at CL=0.5, a typical cruise value.  Here’s what showed up. 

 

Avid_STOL_cruise.thumb.JPG.77f181a4e3d73

 

A bit different looking, eh?  That giant -Cp spike at the leading edge is on the lower surface!  Does the flow stay attached after that steep recovery?  Do we need to VG the lower surface?  Tufts would tell the tale.  I’ll bet some of you have already done this.  Note how much milder it is for the flat-bottomed modification.  By contrast the upper surface has it easy.  So it looks like our friend Manu will have a sweet-flying airplane!

 

One caveat here: these calculations did not include the boundary layer or any of its effects on the outer flow, and were incapable of modelling any flow separation.  In subsonics, everything affects everything, so these results, while indicative of what’s going on, are not accurate to the nth degree.  I also did not include any effects of the flaperon, assuming it was neutral, not lifting upward or downward.  Retired, I don't have a tool available to model the flaperon too.

 

Turbo, i recently bought the pvc leading edge from Kitfox and am working on a template to mount them onto my wings. I downloaded the template that Dholly has posted to this site and imported it into my cad software just to see how accurate it was. The drawing needed to be scaled considerably to match the 2.5" spar and its for a KF4 wing. I'm hoping you would be willing to share your coordinates for the airfoil so I can generate a accurate template and create a pdf that will be much more accurate than the jpeg file.  I will happily post that to this site when I'm done. 

1 person likes this

Share this post


Link to post
Share on other sites

Posted

There are some interesting posts in this thread from a few years ago on extending the leading edge.  May find it interesting.  JImChuk   

www.avidfoxflyers.com/index.php?/topic/4125-leading-edge-extensions/#comment-34447
 

1 person likes this

Share this post


Link to post
Share on other sites

Posted

There are some interesting posts in this thread from a few years ago on extending the leading edge.  May find it interesting.  JImChuk   

www.avidfoxflyers.com/index.php?/topic/4125-leading-edge-extensions/#comment-34447
 

Thanks, i hadn't seen that. I'm even more glad now that I bought it. I know I'm not going to win any races in a KF1 but I do operate at high density altitudes and any increase in climb performance will be welcome. And if i pick up a couple mph and can lower my throttle setting for a lower fuel burn so much the better.

Share this post


Link to post
Share on other sites

Posted

Wilja67, I have only analyzed the Avid STOL airfoil, and don't have coordinates for any KF section.  Is that what you want?

Share this post


Link to post
Share on other sites

Posted

I don't remember where I found it, but I downloaded a copy of Harry Riblett's book, GA Airfoils. This book has the coordinate points for the GA30U-612 airfoil that is used on the Kitfox.

Here are the points in an excel doc and a disorganized cad drawing with those points plotted out. I thought I had a drawing with the Avid undercambered and speed airfoils, as well as the riblett airfoil all plotted out in a cad drawing, but I am not able to find it. 

Riblet Airfoil Plot Points.xlsx

NONAME_0.dwg

1 person likes this

Share this post


Link to post
Share on other sites

Posted

Does anyone know which airfoil is used on the Kitfox STi wing? I found this picture posted on another forum where it shows a normal Kitfox rib in front of a STi rib. It would be interesting to see how the STi airfoil compares to the Avid undercambered wing. I'm guessing the undercambered wing would have more lift at slow speeds but much greater drag at high speeds.

6063.jpg

1 person likes this

Share this post


Link to post
Share on other sites

Posted

Wilja67, I have only analyzed the Avid STOL airfoil, and don't have coordinates for any KF section.  Is that what you want?

Yes,  I was hoping to get the coordinates for the KF 1-3 airfoil. Until recently I assumed the Avid A model and the KF1 airfoil were the same. And I also assumed that the airfoil you analyzed was the same one(looked like it at first glance). I read something somewhere that stated the KF airfoil is a little thicker than the avid? I'm not familiar with Avids since I have concentrated on learning about kitfoxes since that's what I have.

If the Avid airfoil you analyzed is basically the same airfoil just thicker I can scale the thickness in my CAD program to get the correct airfoil. 

Share this post


Link to post
Share on other sites

Posted

I don't remember where I found it, but I downloaded a copy of Harry Riblett's book, GA Airfoils. This book has the coordinate points for the GA30U-612 airfoil that is used on the Kitfox.

Here are the points in an excel doc and a disorganized cad drawing with those points plotted out. I thought I had a drawing with the Avid undercambered and speed airfoils, as well as the riblett airfoil all plotted out in a cad drawing, but I am not able to find it. 

Riblet Airfoil Plot Points.xlsx

NONAME_0.dwg

Although I am hoping to get the early model kitfox airfoil coordinates, this is helpful. 

Share this post


Link to post
Share on other sites

Posted

Doug H posted some pictures a few years ago showing Avid and early Kitfox ribs one on top of the other.  Maybe I'll search for it.  JImChuk

Share this post


Link to post
Share on other sites

Posted

The pictures that Doug posted are inserted and scaled in my cad drawing. The problem though is a picture taken with an uncalibrated camera and lens will have distortion and will affect the accuracy of the measurements taken off the picture. The drawings that Doug posted are very interesting and show the differences between the airfoils but I wouldn't trust them to recreate wing ribs.

Share this post


Link to post
Share on other sites

Posted

The STOL section is a custom 13% thickness airfoil designed originally as a high lift section by Dean Wilson to fit around the 2.5” spar diameter particularly at the leading edge.

The Speedwing section is based on the STOL having more or less similar pressure gradient and mean line but the camber is reduced by approximately 30%.

Both airfoils are therefore specific to the Avid aircraft.

The closest match to the Speedwing section is a NACA 2212 though there are differences.

NACA 44xx series would provide too much pitching moment for the airframe's length and its limited tail volume and should be avoided.

Credit: Graham Laucht

2 people like this

Share this post


Link to post
Share on other sites

Posted (edited)

Doug, I have seen this quotation, but believe it was from David Lednicer, who collects airfoils and runs them through Marc Drela's Xfoil, an Euler-with-boundary-layer code, for comparison's sake.  However, the comment saying the 4-digit NACA sections like 4412, 4415 etc would have too much nose-down pitching moment is simply incorrect.  The Avid STOL section is worse in that regard than those 44xx sections.  Its high cruise drag may be in part profile drag, as all that camber will move the drag bucket out to higher CL, so in cruise we may find ourselves partway up the drag bucket's wall on the low-CL side, but it is also likely that the undercamber, which helps you get off of the runway/lake faster, unnecessarily downloads the tail, increasing induced drag.  My experience with using the roughly 10-deg flaperons on landing approach (steeper descent by quite a bit for only 10 degrees) suggests that the induced drag piece is significant.  This is at fairly high landing CL.  There was not much elevator authority left, so the tail was lifting downward like crazy!  Just a straight-line undersurface would cut the wing's nose-down pitching moment by 40%, but it would take a wee bit more runway to get flying.  Once up, climb rate and angle would be unaffected.

Edited by Turbo
1 person likes this

Share this post


Link to post
Share on other sites

Posted

Dunno. My understanding is that Graham published that info in the UKAVID newsletter. He was involved early on with Dean and Avid, knew and published a lot of Avid Flyer tech info. I believe he may have been a UK dealer as well.

Share this post


Link to post
Share on other sites

Posted

I reckon the attribution doesn't matter much.  My main point is that the NACA 4412, 4415, 4418 all have quarter-chord moment coefficients around - 0.1 in the attached flow region, whereas the Avid STOL airfoil with its lower-surface concavity is closer to -0.14.  This fact is in contrast to the statement in the quotation., which may have been an off-the-cuff, judgment, but is wrong.  Take out that lower-surface concavity, and the AVID shape falls into line with the NACA 44xx shapes.  By the way, the USA35B used, with a minor mod on the supercub and, as I recall, even my old tri-pacer, with its essentially flat bottom, has a CM in the attached-flow range of - you guessed it - -0.1!

Share this post


Link to post
Share on other sites

Posted

My conclusion here is that the STOL airfoil is overcambered, and that a little flap on takeoff would have sufficed for minimizing takeoff roll.  With less camber, the airplane could be more efficient and faster in cruise, with the flaperons pulled back up.  I am told that some folks cruise with flaperons up 3 degs or so, but it looks like they are on the edge of (negative flaperon) stall, while providing downward lift to counter the wing's nose-down moment.  This is not efficient.

I assume you refere to the upper surfice that is overcambered, correct ?

Share this post


Link to post
Share on other sites

Posted

My conclusion here is that the STOL airfoil is overcambered, and that a little flap on takeoff would have sufficed for minimizing takeoff roll.  With less camber, the airplane could be more efficient and faster in cruise, with the flaperons pulled back up.  I am told that some folks cruise with flaperons up 3 degs or so, but it looks like they are on the edge of (negative flaperon) stall, while providing downward lift to counter the wing's nose-down moment.  This is not efficient.

I assume you refere to the upper surfice that is overcambered, correct ?

 

Not necessarily.  It's the undersurface that in my opinion never needed to be concave.  It buys almost nothing, but costs in cruise speed.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

My conclusion here is that the STOL airfoil is overcambered, and that a little flap on takeoff would have sufficed for minimizing takeoff roll.  With less camber, the airplane could be more efficient and faster in cruise, with the flaperons pulled back up.  I am told that some folks cruise with flaperons up 3 degs or so, but it looks like they are on the edge of (negative flaperon) stall, while providing downward lift to counter the wing's nose-down moment.  This is not efficient.

I assume you refere to the upper surfice that is overcambered, correct ?

 

Not necessarily.  It's the undersurface that in my opinion never needed to be concave.  It buys almost nothing, but costs in cruise speed.

A few designers would probably disagree with you on the undercamber statement.

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

And that's totally fair.  If only we had 2-D wind-tunnel data for that airfoil, we could know for sure.  As the sectional CL decreases to zero as the tip is approached, It'd be nice to know that the airfoil isn't climbing up the low-CL wall of its drag bucket.  There's likely a reason that planes like the SuperCub use the essentially flat-bottomed USA35B airfoil.

It's only in this low-speed flight regime that a designer can get away with using one airfoil shape from root to tip.  We sure as hell don't do that in transonic wing design!  It's done for cost reasons on these cheap slow fliers, but to do so successfully, it's important to look at the range of sectional CLs seen at different span stations on the wing in the various phases of flight, and make sure that it never gets too draggy where it counts, like in cruise!  My refined intuition on this (as a retired professional aerodynamicist) says that the undercamber, along with the cylindrical leading edge becomes a liability in cruise, especially outboard near the wingtips.  Then there's the additional nose-down pitching moment, and its effect on tail download and induced drag.  What does the undercamber buy?  A slightly shorter ground run at the AoA set by the gear.  But couldn't the flaperons do this?

Edited by Turbo
2 people like this

Share this post


Link to post
Share on other sites

Posted

So with all this being said.... Have you designed an airfoil that you think would be optimal on these planes to retain the STOL yet be better in cruise?  If so, I am interested in being a test pilot.

:BC:

 

Share this post


Link to post
Share on other sites

Posted

No need.  I think you pegged it with USA35B_mod!

2 people like this

Share this post


Link to post
Share on other sites

Posted

What are your thoughts on the Ribblett airfoils? Do you think the USA35B would perform better than the GA30U-612 that is used on the newer Kitfoxes?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now