Avid STOL airfoil -some calculations

94 posts in this topic

Posted

I have no data on H. Riblett's airfoils, so can render no opinion.  However, computational fluid dynamics (CFD) has come a long way since the USA 35B was designed.  I do tend to believe good wind-tunnel data over CFD predictions, though, and would be hesitant to make comparisons between the two.  CFD is a great design tool, though.

Share this post


Link to post
Share on other sites

Posted (edited)

From what I have read, the USA35B is an old design and there are other more modern airfoils that could perform better on a Cub. I am in the process of modeling the Ribblett GA30U-612 that is used on the newer Kitfoxes in XFLR5 software. I have the wing modeled with an approximation of the standard Avid wingtips. The aerodynamics field is far from my specialty and am finding the learning curve to be steep. My plan is to model and compare the 30U-612, Avid undercambered, and speed ribs to compare the theoretical performance. Maybe I'll add the USA35B to the list as well.

I still have a little ways to go before I get somewhat accurate data, but one thing I found interesting is how much the washout affects the lift toward the wingtips at low angles of attack. The first picture is at 5.5 degree AOA and the second at 1. I have the washout of the wing set at -2 degrees. I'll have to measure the actual washout on Avid wings, but I think I have it close.

612_5_5AOA.PNG

612_1AOA.PNG

Edited by cr125r847
.

Share this post


Link to post
Share on other sites

Posted

Nice going!  I measured 4.5 degrees washout on my Avid C STOL wing, FWIW.

Share this post


Link to post
Share on other sites

Posted

Wow. I didn't know it was that much.

I ran the analysis through an AOA sweep at a fixed lift. With the washout at 2 degrees, the velocity would be 97mph and at 4.5 degrees washout, the velocity would be 119mph! Also the LE of the wing tips do not start contributing to the lift until 7.5 degrees AOA. Interesting...

 

-Brett

Share this post


Link to post
Share on other sites

Posted (edited)

I don't quite understand.  As AoA changes, so does lift.  How can you fix lift while changing AoA?  What velocity are you talking about? 

Yes, the first ~10% chord of that airfoil doesn't appear to produce lift for my cruise case (3kft, 911lbs, 85mph,  wing CL=0.40).  Others defend Dean's choice of airfoil, but I think it has way too much camber.  Why effectively be stuck flying around with the flaps down? (that's what it's like with all that camber!)  For a simple Hershey-bar wing, it's important that the airfoil not be operating at high drag, up on the side of its drag bucket, and we all know the CL falls off going outboard towards the tip.  So the airfoil should work well over the range of CL the wing experiences at all span stations, meaning at low CL too, like out at the tip.  It's much better if the camber level is set appropriate to the important flight conditions - like cruise.  I say the slight shortening of ground run all that camber provides is a bad trade, compared to how it kills cruise.  It would be interesting to see how much faster the bird could be with NACA 23012 on it!  Yes, takeoff ground run would be longer, but by how much?

I think all that washout was intended to protect us hosers from ourselves!

Edited by Turbo
1 person likes this

Share this post


Link to post
Share on other sites

Posted

I keep going back and forth on the airfoils.  In life we know there are theories and realities.  In theory someone could build / design an airplane that will outperform a cub.  In reality, no one has been able to match the performance and guess what's still winning the STOL contests.  Pound for pound, HP for HP, if you want to carry a load and get in and out short (reliably) you go with a PA-18, big tires, VG's and a big flat prop.

Share this post


Link to post
Share on other sites

Posted (edited)

I don't quite understand.  As AoA changes, so does lift.  How can you fix lift while changing AoA?  What velocity are you talking about? 

Yes, the first ~10% chord of that airfoil doesn't appear to produce lift for my cruise case (3kft, 911lbs, 85mph,  wing CL=0.40).  Others defend Dean's choice of airfoil, but I think it has way too much camber.  Why effectively be stuck flying around with the flaps down? (that's what it's like with all that camber!)  For a simple Hershey-bar wing, it's important that the airfoil not be operating at high drag, up on the side of its drag bucket, and we all know the CL falls off going outboard towards the tip.  So the airfoil should work well over the range of CL the wing experiences at all span stations, meaning at low CL too, like out at the tip.  It's much better if the camber level is set appropriate to the important flight conditions - like cruise.  I say the slight shortening of ground run all that camber provides is a bad trade, compared to how it kills cruise.  It would be interesting to see how much faster the bird could be with NACA 23012 on it!  Yes, takeoff ground run would be longer, but by how much?

I think all that washout was intended to protect us hosers from ourselves!

I think the reference on the AOA is trying to get the wing tips flying.  All you have to do is look out the window in cruise and see how far down that leading edge of the tip is.  There is no way possible that LOADS of drag are not present.  I am not an engineer nor do I  have a PHD in fluid dynamics, but I have stuck my hand out the car window more than once and I know what happens when you tilt your hand.  

:BC:

 

Edited by akflyer
1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

My first line, last post was for CR127, who put up some impressive graphics.  That discourse seemed too long to quote.

Leni, I think your intuition serves you well.  I expect Scrappy will show everybody up, but like Mike Patey says, it's not a practical airplane.  My Cp plots on the Avid wing (next thread, unfortunately, as the site wouldn't  let me post my plots as a new post inside this thread!) show the airflow at cruise grossly oversped on the lower surface outboard near the tip and just behind the leading edge.  The rapid deceleration just downstream of the leading edge might just separate the flow on the lower surface; we can't know without more calculation, or even better, wind-tunnel testing.  Tufting that area of the wing would tell the tale, and it's visible from the cockpit, being on the underside.  But that slowdown and pressure-rise for sure beats the sh#@$t out of the poor boundary layer!

The tried and true USA35B_mod will serve you well, I think.  Fancy new section shapes are out there, but unless you can compare apples-to-apples, i.e. wind-tunnel data to wind-tunnel data (not wind-tunnel data to computation), and are sure you know which will be better for your specific conditions, I'd stick with tried-and-true.  Building or rebuilding a wing is a lot of work!

So what's my read on John Monnett's choice of NACA 64_415 on the Sonex?  Too much camber!  With a different airfoil choice, even that rockin' little bird could be faster!

Edited by Turbo

Share this post


Link to post
Share on other sites

Posted

I don't quite understand.  As AoA changes, so does lift.  How can you fix lift while changing AoA?  What velocity are you talking about? 

Yes, the first ~10% chord of that airfoil doesn't appear to produce lift for my cruise case (3kft, 911lbs, 85mph,  wing CL=0.40).  Others defend Dean's choice of airfoil, but I think it has way too much camber.  Why effectively be stuck flying around with the flaps down? (that's what it's like with all that camber!)  For a simple Hershey-bar wing, it's important that the airfoil not be operating at high drag, up on the side of its drag bucket, and we all know the CL falls off going outboard towards the tip.  So the airfoil should work well over the range of CL the wing experiences at all span stations, meaning at low CL too, like out at the tip.  It's much better if the camber level is set appropriate to the important flight conditions - like cruise.  I say the slight shortening of ground run all that camber provides is a bad trade, compared to how it kills cruise.  It would be interesting to see how much faster the bird could be with NACA 23012 on it!  Yes, takeoff ground run would be longer, but by how much?

I think all that washout was intended to protect us hosers from ourselves!

Lift changes with the AoA at a constant airspeed, but the analysis I did was an AoA a sweep at fixed lift (750 lbs in this case). The variable is airspeed.

What I'm after with the simulations is to compare the different airfoils and see what effect changing from the Avid undercambered airfoil will have while leaving the wing incidence as it is. There was talk a few years ago about how Kitfox increased the angle of incidence when they switched to the Ribblett 612. Would this need to be changed to make an Avid perform well with the 612 airfoil?

It would also be fun to compare the Kitfox STi rib to see where it falls. Does anyone know what airfoil they are using?

Share this post


Link to post
Share on other sites

Posted

I keep going back and forth on the airfoils.  In life we know there are theories and realities.  In theory someone could build / design an airplane that will outperform a cub.  In reality, no one has been able to match the performance and guess what's still winning the STOL contests.  Pound for pound, HP for HP, if you want to carry a load and get in and out short (reliably) you go with a PA-18, big tires, VG's and a big flat prop.

Agreed and I didn't intend to sway this into a "Cub Killer" conversation. I don't think a Kitfox will ever be able to compete with a Cub due to the narrow, high AR wing on the Kitfox. One of the compromises of having a foldable wing.

Share this post


Link to post
Share on other sites

Posted

My first line, last post was for CR127, who put up some impressive graphics.  That discourse seemed too long to quote.

Leni, I think your intuition serves you well.  I expect Scrappy will show everybody up, but like Mike Patey says, it's not a practical airplane.  My Cp plots on the Avid wing (next thread, unfortunately, as the site wouldn't  let me post my plots as a new post inside this thread!) show the airflow at cruise grossly oversped on the lower surface outboard near the tip and just behind the leading edge.  The rapid deceleration just downstream of the leading edge might just separate the flow on the lower surface; we can't know without more calculation, or even better, wind-tunnel testing.  Tufting that area of the wing would tell the tale, and it's visible from the cockpit, being on the underside.  But that slowdown and pressure-rise for sure beats the sh#@$t out of the poor boundary layer!

The tried and true USA35B_mod will serve you well, I think.  Fancy new section shapes are out there, but unless you can compare apples-to-apples, i.e. wind-tunnel data to wind-tunnel data (not wind-tunnel data to computation), and are sure you know which will be better for your specific conditions, I'd stick with tried-and-true.  Building or rebuilding a wing is a lot of work!

So what's my read on John Monnett's choice of NACA 64_415 on the Sonex?  Too much camber!  With a different airfoil choice, even that rockin' little bird could be faster!

When you hit the reply button it should still have same drag and drop options to attach files or click the choose file button.  

Share this post


Link to post
Share on other sites

Posted

I don't quite understand.  As AoA changes, so does lift.  How can you fix lift while changing AoA?  What velocity are you talking about? 

Yes, the first ~10% chord of that airfoil doesn't appear to produce lift for my cruise case (3kft, 911lbs, 85mph,  wing CL=0.40).  Others defend Dean's choice of airfoil, but I think it has way too much camber.  Why effectively be stuck flying around with the flaps down? (that's what it's like with all that camber!)  For a simple Hershey-bar wing, it's important that the airfoil not be operating at high drag, up on the side of its drag bucket, and we all know the CL falls off going outboard towards the tip.  So the airfoil should work well over the range of CL the wing experiences at all span stations, meaning at low CL too, like out at the tip.  It's much better if the camber level is set appropriate to the important flight conditions - like cruise.  I say the slight shortening of ground run all that camber provides is a bad trade, compared to how it kills cruise.  It would be interesting to see how much faster the bird could be with NACA 23012 on it!  Yes, takeoff ground run would be longer, but by how much?

I think all that washout was intended to protect us hosers from ourselves!

I think the reference on the AOA is trying to get the wing tips flying.  All you have to do is look out the window in cruise and see how far down that leading edge of the tip is.  There is no way possible that LOADS of drag are not present.  I am not an engineer nor do I  have a PHD in fluid dynamics, but I have stuck my hand out the car window more than once and I know what happens when you tilt your hand.  

:BC:

 

I used to stick my hand out the window in the truck, till a dang bumblebee center punched it.  Yes, it hurts! Probably hurt the bee more, though.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

The cruise AoA of the fuselage is not that important, within bounds, of course.  More important might be retaining the ability to fold the wings,   That may be what drove the incidence change when going to the new airfoil.  But be careful when considering new "higher tech" airfoils.  Design assumptions could trip you up in practice.  Here's an example:

Years ago a colleague of mine used the then-latest boundary layer theory and even invoked variational methods to determine the optimum shape of the Cp distribution for maximizing single-element lift, and achieved a sectional CL of about 2.2 at 5 million Reynolds number, more-or-less where we fly.  Homebuilders were calling him regularly asking if his airfoil was appropriate, and if they could use it on their airplanes.  With CL max of 2.2, man, could you shrink the wings (in principle, anyway!).  His standard answer was "try NACA4415."  This guy was well known in the aerodynamics world, and became a senior technical fellow of the corporation.  His airfoil was tested in a low-turbulence wind tunnel and performed substantially as predicted.  Of course, being optimized for high lift, it had super-nasty stall characteristics, and really fell out of the sky when it let go.  (Can we say "snap-roll", boys and girls?) There were other considerations that amateur armchair aerodynamicists might be hard-pressed to understand, that had to do with surface finish and the need (and technology) to carefully orchestrate the boundary layer state (laminar vs. turbulent).  Those and other considerations made his airfoil poorly suited for general application to experimental aircraft.  NACA 4415, however, has test data over a wide range of operating conditions, and is relatively immune to small surface roughness (say, due to fabric weave) and boundary layer state.  It just works, reliably.  Oddly, with its essentially flat bottom, it looks a lot like USA35B.

I spent most of my career not doing CFD, but instead developing the technology for keeping the boundary layer in the low-friction laminar state on large, swept, transonic, and even supersonic wings.  This culminated in the design of the successful active laminar-flow system on the production Boeing 787 -9 and -10 tailfeathers, after which I retired.  A rule of thumb I use is: if there's distributed tactible roughness, like fabric weave, assume the boundary layer is turbulent.  This has certainly worked on my windsurf fins.  All things equal, ones with a matte finish tend to work better than perfectly smooth ones.  Laminar boundary layers are notably wimpy.  Turbulent ones are way more robust in the face of rising pressures in the downstream direction.

2 people like this

Share this post


Link to post
Share on other sites

Posted

Fist paragraph of my last post is o.k., except for the prop's effects.  The propshaft, if at AoA, will produce a yawing moment, which would require a rudder trimtab to null out, so there's another reason for re-setting wing incidence when going to a different airfoil shape, with its different zero-lift AoA.  Cruise altitude, speed, and weight play into this, so the designer has to pick some representative cruise point for setting wing incidence.

Share this post


Link to post
Share on other sites

Posted

I agree.  Even a straight-line, flat underside will work fine.  I sure wish there was an easy way to put detents on the flaperon handle.  I've got to at least upgrade my friction discs.

Hey Turbo, I 3d printed a flap handle detent setup, details in this post

http://avidfoxflyers.com/index.php?/topic/6873-3d-printed-flap-detent/#comment-62183

Ken

And a brilliant design it is!  Kudos to you!  Unfortunately, in my case, Elvis has left the building.  I am now the proud owner of a Corvair-powered Sonex!

Share this post


Link to post
Share on other sites

Posted

I see a lot of discussion but has anyone removed some of the undercamber and flown their Kitfox or Avid Flyer? I am almost at the wing cover stage and would remove some of the undercamber if I knew the true tradeoff. I would like another 8-10 MPH in my Kitfox II as long as the takeoff roll doesn't double.

Share this post


Link to post
Share on other sites

Posted

There's a European guy, Manu, I think, (Efil01 ??) who has done that.  Whether he's flying yet or not I can't say.  You would pay a small penalty in takeoff distance, but rolling in a little flaperon deflection for takeoff, 5 or so degrees, ought to get you back to equivalency with the old wing, but no flaperon, as would bigger tires (taller mains).  Its only effect is seen when you're AoA limited while the gear is still on the runway.  So if it were me, I'd go for it!  With all that wing twist, the drag ought to be reduced significantly by filling in that concave lower surface.  Less drag means more speed.

1 person likes this

Share this post


Link to post
Share on other sites

Posted

There's a European guy, Manu, I think, (Efil01 ??) who has done that.  Whether he's flying yet or not I can't say.  You would pay a small penalty in takeoff distance, but rolling in a little flaperon deflection for takeoff, 5 or so degrees, ought to get you back to equivalency with the old wing, but no flaperon, as would bigger tires (taller mains).  Its only effect is seen when you're AoA limited while the gear is still on the runway.  So if it were me, I'd go for it!  With all that wing twist, the drag ought to be reduced significantly by filling in that conc

 hi turbo, yes i fly, have sent you e mail... i'll be pleased to talk about my mod  and these numbers

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