Stall progression

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,420
Location
North Carolina
Some of you may have noticed that I'm a little wary of designing a wing from scratch and getting it's stall characteristics acceptable. Half the trick is getting the inboard portion to stall first. Here's a great NASA paper about doing just that. Some of my Hurricane aero efforts will be to maintain it's reasonable stall characteristics at my lower Re.
It isn't all airfoils. The Fw190 had a famously vicious stall. At high speed. At low speed, it was apparently fairly tame. It's problem was aeroelasticity. Under high g, the wing would wash in. When it stalled, it stalled a tip...
 

Arthur Brown

Well-Known Member
Joined
May 1, 2016
Messages
138
Location
London
The DH Chipmunk initially had a vicious stall with the tips stalling first so they fixed little triangular sections to the inboard front of the wing leading edge, to force the roots to stall first -with a more gentle effect. So stall behaviour can be treated afterwards.
From memory they were 1" by 1" and 20 to 30 inches long with a leading edge radius of 0.125"
 

Jan_K

Member
Joined
Sep 3, 2014
Messages
13
Location
Czech Republic
You have at least two feasible options:
- stall center section first
e.g. Zlins 26 line had 4mm wire, about 50cm long on the leading edge
The wire can be held in place just by a strip of fabric + clear dope.
not very well visible on photographs
http://album.medlanky.info/albums/userpics/normal_100_3648.JPG
https://imgproc.airliners.net/photos/airliners/9/3/2/2006239.jpg?v=v40

- vortex generators
Installation of Vortex generators behind leading edge in the aileron span would improve stall behavior and also aileron response.
I have very positive experience with this, full span VGs had reduced stall speed of a microlight from 64 to 45km/h (forty five).
find out more details here
https://www.stolspeed.com/id/60
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
Lots of options for minimizing the effect. The FW 190 had the centroid of the wing well back and that combined with high g's would warp the tips down under positive g and up at negative g. This can be all calculated by hand or with FEA. If you work out the net moment about the centroid, and work out the torsional stiffness, you can figure out how much the wing washes-in at max g. If you have "too much twist", there are solutions at the design level then after you are flying it.

Let's start with either some data or calcs on an airplane with a too limber wings, like the FW190. Do you know how much wash-in they had at what g's? Maybe calculate from structural details how much twist they could get. Then with that info as a chinning bar, check your wing. Twist is the sum of pitching moment (which unloads the tips) plus lift (at quarter chord) times arm (from quarter chord back to wing centroid). On to design level:

You can move the centroid forward to reduce the twisting moment;
Increase torsional stiffness of the whole wing;
Design in some washout;
Design the aileron linkage to gently reflex the ailerons as the wings bend and twist.

Once built and flying, you move to aerodynamic fixes. Stall strips and wedges inboard, vortex generators outboard and the little gadgets the Glasair line uses, drooped leading edges outboard.

I would urge you to be smart. First find out how much twist is too much by analyzing "bad" wings then comparing to yours. Maybe even analyze a couple wings that are acceptable too, to give you an idea of how good you need to be. See what you can do design-wise with your wing and how much weight penalty it gives you. What is left when you get to flight test, you handle with aero fixes. If instead you just accept the design with a too much twist per g, you may have the devil's own time getting it to behave well.

I figured out that I had 0.4 degrees of wash-in with my wing at 6 g, and so put in 0.4 degrees of wash-out at construction. That combined with a Riblett airfoil and following Harry's thoughts on wing taper and speed strakes should leave little problem with stall behaviour. VG's and stall strips can be added if needed.

Billski
 

Jan_K

Member
Joined
Sep 3, 2014
Messages
13
Location
Czech Republic
Well, the wing geometry plays important role, not just airfoil shape. The methods described above are suitable for mending troubles of an existing wing.
If you are designing wing from scratch, I would suggest to google first for wing design using Prantl-Glauert's method.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,900
Location
Warren, VT USA
You can also use multiple airfoils that have different stall characteristics inboard and outboard and blend the geometry in your wing design. You want an airfoil at the tips that hangs on at an AOA that root stalls. This can be a better lower drag solution vs lots of twist. Twist is a way to get a good overall lift distribution out of a rectangular wing. Forward sweep and taper can cant the trailing edge forward leaving the leading edge perpendicular to the thrust line. This config favors root first stall.

I hate seeing patches on a brand newly designed wing. Not necessary.

Love Billski's description of washout per G by manipulating the centroid. Good stuff. In composite wings the angle of laminations and the centroid is a whole world of opportunities to optimize a wing for flutter, twist, etc...
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,005
Location
USA.
At first my Falconar F-12 had a very sharp stall and would drop in a spin in either direction, very easy to stop the spin. I always kept extra speed in the pattern in the turns. Falconar came out with LE stall strip to fix the problem. Just a 1/2" triangle strip of wood about 8" long on the LE of the wing inboard of the landing gear about 20" from the side of the fuselage. Huge difference. Then stalled like a Piper Cherokee, but cost 200 FPM in the ROC. 1700 fpm at GW to 1500 fpm.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
Would making the inboard leading edges a little sharper than the tip leading edges do the trick safely?
Who says pictsidhe needs to fix what he has? If he has started with a foil with benign characteristics, a modestly tapered plan form, decent torsional stiffness and centroid position, and carries the same foil from root to tip, it may already stall just fine from 1 g on up to his g limit. To start including fixes before you even know is silly. I urge doing the analysis up front, including looking at other airplanes with known characteristics before you get wrapped around the axle.

As to specifically sharpening the foil near the root and smoothly transitioning to a fatter nose, I do not like it. Besides the build complexity, if you sharpen the foil when it is not needed, then you have a less desirable foil over much of the wing. If you find in flight test that you do need to force a stall towards the root, you can add stall strips where you need them later, tune with them, and then make permanent the ones that work for you... Likewise if you find you need to delay stall outboard, you can add and tune wedges, VG's, or even the neat gadget the Glasair products use. Drag contribution of all these fixes is measurable so only use the ones you can show that do you some good.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
At first my Falconar F-12 had a very sharp stall and would drop in a spin in either direction, very easy to stop the spin. I always kept extra speed in the pattern in the turns. Falconar came out with LE stall strip to fix the problem. Just a 1/2" triangle strip of wood about 8" long on the LE of the wing inboard of the landing gear about 20" from the side of the fuselage. Huge difference. Then stalled like a Piper Cherokee, but cost 200 FPM in the ROC. 1700 fpm at GW to 1500 fpm.
Sounds like Falconar went for docile and added a bunch of drag. Maybe the strips needed to be that size, maybe they erred on the side of caution.

The stall strip on my wife's RANS S6 is much smaller, smoothly radiused and fairly thin, and is just glued to the forward spar tube. Being as the design is fairly well developed, the fabric goes over it. Docile stall.

Some certified metal birds have stall strips 3/8" wide and only a few inches long, attached with a couple screws or rivets.

Billski
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,005
Location
USA.
The stall strips did not slow the airplane down, 150 mph cruise with a climb prop. Just effected the ROC. Really, I like it better without the stall strips after flying it 5 years without the strips.
Yes, I guess Falconar wanted to go to docile and error on the side of caution. I have flown about 85 different airplanes and the F-12 is my favorite. Wife and I did a lot of traveling in it over the years. Great traveling machine.
Look at some of the Piper Arrows, 3 stall strips on each wing. And they still ended up with a poor airplane.
 

Riggerrob

Well-Known Member
Joined
Sep 9, 2014
Messages
1,403
Location
Canada
Thanks for filling in that gap about aeroelastic wings on FW-190.

Many of the problems - with tip stall on small airplanes - are caused by tiny Reynolds.numbers. At tiny Reynolds numbers, even the tiniest of imperfections can trip airflow from laminar to turbulent.
That is why it can be counter-productive to design tip chords less than 3 or 4 feet. At that size, there are few weight savings, but structure gets too complicated.
Besides, simple rectangular wings have almost elliptical lift curves and naturally stall at the roots first.

As for the OP’s question ....... look at the Quest Kodiak’s extended leading edges on the outer wings. Those extended leading edges enlarge the leading edge radius (ala. Robertson, Sportsman STOL, etc.) helping smooth airflow over ailerons even when the roots are stalled.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,420
Location
North Carolina
I didn't post this as a question!
I am scaling down the Hawker Hurricane to 103 size and speed. There is a little redesigning taking place to achieve that. The full size plane has acceptable stall characteristics that I don't want to lose after my 'minor' changes. That paper was a great find, and I thought others may be interested.
My plan is to find a pair of airfoils that stall the wing at least as benignly as the full size ones. If you plough through that paper you'll see that the 4 and 5 digit wings lose stall margin as Re decreases. Exactly what I was afraid of! That paper also shows that my 0.5 taper is down in the just feasible range... The original airfoils on the hurricane are Clark YH 12 and 19%, which probably behave a lot like 4 digit NACAs. Now, I just need to find a suitable pair to work at my Re and I should be good. Assuming that I can avoid aeroelastic effects biting me instead...
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I'm speculating, but I'm guessing that pictsidhe wants to know that the stall characteristics are right before the aircraft takes off with him inside it.

Since his project is very light, two approaches come to mind, though they do have logistical problems:

1- find someplace to get away with testing by RC. It would probably be good to take some data, not just relying on how it feels to an RC pilot. In particular, stick forces. I understand that when the Facetmobile was first tested, the stick would tend to pull back more when pulling up elevator. This wouldn't be revealed by a radio controlled test that didn't measure stick forces.

2. Do something with a truck mount. Hang gliders have been tested with trucks and the BD-5 had a training rig attached to a truck. Come to think of it, a part 103 ultralight could be structurally tested with a truck as well, to many g's.

The 2nd approach could be tough on the budget. The 1st one could as well if much travel or lawyer-wrestling was involved. A sub-scale RC test might provide SOME useful information, but of course the Reynolds numbers would now be too low.

I suppose that one could do a sub-scale test in water, but a large boat would probably be required. You could get by with just a rope tow and some kind of remote control. However, the forces involved would be much higher than testing in air, even with a 1/3 scale model at an equivalent Reynolds number.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,420
Location
North Carolina
Yes lr27, I'm trying to fix the bad stall brought about by directly scaling a WWII fighter to 103 before I build it, let alone fly it! I do like the idea of truck testing, but I don't know of any good places to do it in my neck of the woods. I'll look into sailplanes.

It's getting annoying having to redesign the ailerons when I change the tread pattern on the tailwheel, but I am making progress. Ok, maybe that's a slight exaggeration, I haven't changed the tailwheel tread yet!
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
13,084
Location
Port Townsend WA
The NASA cuff paper is more recent.
Outboard cuffs even work on RC model scale Reynolds numbers.
 

Marc W

Well-Known Member
Joined
Mar 31, 2017
Messages
420
Location
Colorado
I have to ask! You will have low wing loading and low speed. How bad can an ultralight stall be? Maybe you are over thinking this?
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,694
Location
Saline Michigan
The stall strips did not slow the airplane down, 150 mph cruise with a climb prop. Just effected the ROC. Really, I like it better without the stall strips after flying it 5 years without the strips.
Yes, I guess Falconar wanted to go to docile and error on the side of caution. I have flown about 85 different airplanes and the F-12 is my favorite. Wife and I did a lot of traveling in it over the years. Great traveling machine.
Look at some of the Piper Arrows, 3 stall strips on each wing. And they still ended up with a poor airplane.
Pops, it sounds like a great airplane. I gotta believe you know its performance in both configurations...

If the climb rate went down after adding the stall strips and you are putting in power exactly the same (same prop and engine and airplane weight and climb speed) something is using up more power. That something is usually more drag, even if it only affects things at climb AOA. Maybe the stall strips were tripping the boundary layer and adding some drag during climb that they were not tripping in cruise? Maybe you climb at a higher airspeed with the stall strips?

Billski
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,005
Location
USA.
I think you might be correct. Same speed before and after installing the strips as far as I could tell and I few that airplane a lot all over the U.S. But without the strips at GW I had a ROC of 1700 fpm and with just me and about 1/2 fuel tank the ROC was 2200 fpm. Both went down 200 fpm with the insulation of the stall strips. Also lost the fun of the stall :)
Like you say, weight is the enemy, EW was just 937 lbs.
It climbed so good, to stay down and climb with a normal rate as most Cessna and Pipers, I would be climbing out at 110-120 mph.
One of the biggest mistakes I have made is selling the F-12. Always regretted selling it. One owner ground looped it and repaired the damage and sold it and the next owner destroyed it in a ground loop and it was a very easy taildragger, but like all taildraggers, you have to stay awake.
 
2
Top