Quantcast

Tip stall and highly tapered wings

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
I found a couple of papers that seem (to me) to give a pretty good "picture in the mind" explanation of why highly tapered wings have tip stall problems. From what I can discern (caveat), it appears that, to put it simply, the area of a highly tapered wing toward the tip becomes progressively "too small" to carry its fair share of the spanwise lift requirement even with the effects of outward boundary layer flow effectively decreasing (edit: wrong, see Norman's post) the angle of attack, tip vortices, and washout. No doubt it's more complex but this seems to be the basic gist.
The MIT paper also suggests that although Reynolds numbers are not the cause of this phenomenon, they do exacerbate it and I wonder if that is why the tip stall problem is more acute for smaller aircraft?
Pic 1 is the MIT paper and pics two and three are also a good (and simple)explanation of the problem as it applies to VERY small aircraft:
 

Attachments

Last edited:

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
I found a couple of papers that seem (to me) to give a pretty good "picture in the mind" explanation of why highly tapered wings have tip stall problems. From what I can discern (caveat), it appears that, to put it simply, the area of a highly tapered wing toward the tip becomes progressively "too small" to carry its fair share of the spanwise lift requirement even with the effects of outward boundary layer flow effectively decreasing the angle of attack, tip vortices, and washout. No doubt it's more complex but this seems to be the basic gist.
The MIT paper also suggests that although Reynolds numbers are not the cause of this phenomenon, they do exacerbate it and I wonder if that is why the tip stall problem is more acute for smaller aircraft?
Pic 1 is the MIT paper and pics two and three are also a good (and simple)explanation of the problem as it applies to VERY small aircraft:
It has more to do with where the stall starts and the direction it moves across the wing surface. Like this:



On anything other than the plain rectangular wing, washout is usually needed to control stall behavior. Stall strips are often necessary, too.

Dan
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,208
Location
Orange County, California
And of course, here's one of the places where the compromises start. You want an elliptical distribution for low induced drag (especially for a good climb rate or in hard turns), but an elliptical distribution tends to 'let go' all at once in the stall - an undesireable trait. Washout is just another tool to change the lift distribution.
 

HumanPoweredDesigner

Well-Known Member
Joined
Sep 6, 2009
Messages
1,030
Location
Arizona
How common are tip stalls? Can't you just use a rounder airfoil near the tips so it has a better range of AoA before stall? Or use one at the tip that stalls at a higher AoA? Then it would have lower coefficient of lift during normal flight and add to the taper effect.
 

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
It has more to do with where the stall starts and the direction it moves across the wing surface. Like this:
Umm..., and the stall starts at a certain point on the wing because the coefficient of lift at that point is higher than at other points on the wing?
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,469
How common are tip stalls? Can't you just use a rounder airfoil near the tips so it has a better range of AoA before stall? Or use one at the tip that stalls at a higher AoA? Then it would have lower coefficient of lift during normal flight and add to the taper effect.
It's commonly done. See it on the wing of the Quest Kodiak:




Dan
 

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
How common are tip stalls? Can't you just use a rounder airfoil near the tips so it has a better range of AoA before stall? Or use one at the tip that stalls at a higher AoA? Then it would have lower coefficient of lift during normal flight and add to the taper effect.
The rounder airfoil near the tip does a similar thing to what washout does, i.e., uh.. it allows the tip to keep "flying" a little after the rest of the wing has stalled? And uh... at some point these aerodynamic "compensations" like washout, higher lift sections, slots or slats can no longer "compensate" for the highly tapered planforms propensity to stall tip first? I think..., maybe.:ponder:
(Hopfully someone else will give a real answer:gig:).
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,135
Location
Saline Michigan
Well, severe taper is not a great idea, but there is more going on than just taper...

All of these articles and diagrams come from the original NACA series of foils and plan form tests that they did back then. Harry Riblett is highly critical of several processes here, and I buy some of his arguments. I will explain:

First off, all of the classic tapered wing testing was done with wings that not only tapered the chord, they also used foils at the tip that were of the same family but with % thicknesses lower than at the root. This was thought to be efficient wing design at the time, but it had unintended consequences. The NACA foils used a way of drawing the leading edge that actually reflexes the leading edge a bit (making the stall sharp) and results in the thinner foils having somewhat lower AOA at stall... So the tests ended up confounding two issues - wing taper and airfoil choice at root vs tip. Did they report that they had confounded their data? Nope. Did they attempt to separate the issues with tapered foils using the same foil root to tip? Nope. They just attributed the poor stall progression to tapered and swept wing shapes without mention of the foils also contributing. Only the rectangular wings and the elliptical wing used the same foil... Oh, but isn't the elliptical foil a tapered foil? And it behaved OK. Hmmmm.

It turns out that you can taper a wing using the same foil from root to tip and have very nice stall behaviour. Look at the Piper Commanche (both single and twin versions) as an example.

In addition, Harry's foils actually build in a bit of camber line droop at the leading edge, which soften's the stall. And Harry also advocates modest taper - no more than 50%.

Since then, a number of different airplanes have been built using Harry's approach, and what do you know, they don't exhibit sharp stall or tip stalls...

Then there are the folks (Orion among them) who feel that you should tailor the local Cl to get nice benign stall behaviour and an elliptical lift distribution by tailoring the taper and the foils used... Cool stuff. Efficient wings. Good stall behaviour.

So please don't just buy the old saw about tapered wings being prone to tip stalls. They don't have to be that way...

Billski
 

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
From the looks of the lift curve on airfoil charts it would seem that a particular airfoil will always stall at the same AOA give a consistent Reynolds number (and I've heard that said on this forum, as well). This is weird. Is Reynolds number effect (not overall, but as it varies from root to tip) at the root of this phenomena?
 

Autodidact

Well-Known Member
Joined
Oct 21, 2009
Messages
4,513
Location
Oklahoma
Its almost more like a Mantra than a saw - "Tapered wings are prone to tip stall,tapered wings are prone to tip stall, tapered wings are prone to tip stall, tapered wings are......):D
 

Norman

Well-Known Member
HBA Supporter
Joined
Nov 28, 2003
Messages
3,043
Location
Grand Junction, Colorado
FIs Reynolds number effect (not overall, but as it varies from root to tip) at the root of this phenomena?
No... The root cause is induced upwash that causes sections farther outboard to see a higher AoA than the root. Both taper and sweep do this. The Weissinger computation that I pointed to in the first reply calculates this correctly. This is more accurate than the lifting line theory that was the best they had when NACA was doing those experiments that Billski is so down on.
 

HumanPoweredDesigner

Well-Known Member
Joined
Sep 6, 2009
Messages
1,030
Location
Arizona
Umm..., and the stall starts at a certain point on the wing because the coefficient of lift at that point is higher than at other points on the wing?
No, not the coefficient of lift. One airfoil can be at twice the coefficient of lift of the other and still have 7 degrees to go before it stalls. Just pick the airfoils right. Some have good L/D through a range, and good Cl through a range. Just pick your tips so that that they stall at a higher AoA. Only problem is if you nose down suddenly, your tips will stall first then too. I'd go with a rounder leading edge for the wing tips.
 

HumanPoweredDesigner

Well-Known Member
Joined
Sep 6, 2009
Messages
1,030
Location
Arizona
No... The root cause is induced upwash that causes sections farther outboard to see a higher AoA than the root. Both taper and sweep do this. The Weissinger computation that I pointed to in the first reply calculates this correctly. This is more accurate than the lifting line theory that was the best they had when NACA was doing those experiments that Billski is so down on.
What is induced upwash? The outboard flowing air under the wing? How does that make the outboard wing see a greater AoA?
 

pie_row

Well-Known Member
Joined
Nov 9, 2009
Messages
710
Location
salt lake city Ut
if you have taper then the out board area doesn't have the oomph to keep up with the indoard area. With a 4:1 taper ratio the inbord area has 4*the power that the out board area has and so you get tip stalls really bad.
 

Michealvalentinsmith

Well-Known Member
Joined
Oct 13, 2009
Messages
361
Location
Rainbow beach QLD Australia
Apparently the tip stall and spin is the single biggest factor in many aviation accidents. A common scenario is the pilot slows down to enter the pattern, turns 90 degrees tip stalls and spins in with insufficient altitude to recover. It started with the Wright brothers - who went with canards as a "solution" (at least it provided crash protection) and still occurs.

I recall reading a book in the 80s where a guy went all out for a solution. He came up with slats, slots and spoilers coupled with ailerons. On the up-going side a spoiler popped up behind the slat to close the the slot, as the aileron went up. On the down going side the slot stayed open keeping the flow attached at higher AofAs.

The author claims he eliminated adverse yaw and had full control at, and through stall. It sounds draggy and complex to me but it did seem to work. Later I figured why not just rotate the slats to act like spoilers? I thought I had an invention but found out later it has been tested by the NACA and worked reasonably well but the yaw moment is larger than the rolling moment. Apparently early Junkers aircraft had slats that were opened by suction. If they opened unevenly a nasty roll occurred and this is where the idea started.

Regarding stall patterns I read that even a rectangular planform can still stall at the tips depending on the airfoil used and turbulence of the airflow- I sure found this in RC wings but that's probably an unrepresentative example. Also the elliptical wing spitfire had nasty stall characteristics, and the elliptical planform is lost as soon as you deflect an aileron. Most guys achieve a near elliptical planform with taper and twist these days.

I read a report that claimed a crescent shaped wing is actually more efficient than an elliptical in real world tests. I can't recall why now.
 

Michealvalentinsmith

Well-Known Member
Joined
Oct 13, 2009
Messages
361
Location
Rainbow beach QLD Australia
Regarding why the tips stall with high taper I suspect it has something to do with loading as well. The higher the taper the higher the loading. If you think of the tips as separate sections a more highly loaded wing wing stall at a higher stall speed.

When paraglider wings collapse in turbulence it's not uncommon for the pilots to accidentally negatively spin the residual portion of the wing with control application as the remaining wing has a much higher stall speed.
 

Topaz

Super Moderator
Staff member
Log Member
Joined
Jul 29, 2005
Messages
14,208
Location
Orange County, California
Umm..., and the stall starts at a certain point on the wing because the coefficient of lift at that point is higher than at other points on the wing?
Effectively, yes. A combination of induced-upwash and finite-span effects creates a lift coefficient distribution (and so a lift distribution) that varies across the semi-span. When the region of the wing that is carrying the highest lift coefficient reaches the stall angle locally, that point of the wing begins to stall. The stall spreads as other regions reach their critical angle locally.

The rectangular wing tends to load up the root more quickly, whereas a wing with strong taper tends to load up the tips. An elliptically-loaded wing will have a constant lift coefficient distribution across the span - which is why they tend to stall the whole wing at once.

As Billski pointed out, it's not a matter of 'tapered wings will have tip stall'. Using washout (geometric twist), differing airfoils (aerodynamic twist); etc., you can tailor the lift distribution (and the stall behavior) of practically any arbitrary wing planform. There is NO simple 'rule of thumb' for this process once you leave a rectangular planform. You just have to figure out the lift and lift coefficient distributions and determine the stall behavior from there.
 
Top