stall!

[propstar]

witch would stall first on an oldy slowfly with a flat bottom wing -thick or thin.

[J D 8 - Moderator]

Probably the thick wing because of more drag but likely not much in it I recon [ just my thoughts] Slow fly's do they stall at all or just become a falling leaf?

[propstar]

cheers mod,tend to agree,pehaps a tad of washout on either.

[Chris Walby]

But would it...agreed it produces more lift so more effective at low speed...does it depend on the AOA and if a thin wing airflow breaks down first?

[Don Fry]

I think, a thick wing goes first, but it warns you, thin goes viciously, little warning. A long, long time ago, I flew pylon racers, and the turns, as you drift into a competitor, and he’s not giving an inch, does not end well. 

[Simon Chaddock]

My understanding of a stall is when the lamina airflow breaks away from the upper wing surface and becomes turbulent at a angle of incidence that depends on the wing section. This reduces lift, increases drag and alters the position of the centre of lift.

The relatively sharp leading edge radius of a thin wing would promote the break away before the blunt radius of a thicker wing but I could be wrong.

How a wing behaves once the stall is under way is a much more complex situation and can even favour wing sections and shapes that actually promote a flow break away.

[Chris Freeman 3]

Should be the thin wing, A thick wing is higher lift but also in drag. 

[john stones 1 - Moderator]

Thin wing, the funflys have thick sections, stalls nigh non existant.

[PatMc]

14 hours ago, propstar said:

witch would stall first on an oldy slowfly with a flat bottom wing -thick or thin.

It depends on whether she's on her heavy broom outdoor broom her light indoor one. 😁

[Peter Jenkins]

19 hours ago, Simon Chaddock said:

My understanding of a stall is when the lamina airflow breaks away from the upper wing surface and becomes turbulent at a angle of incidence that depends on the wing section. This reduces lift, increases drag and alters the position of the centre of lift.

The relatively sharp leading edge radius of a thin wing would promote the break away before the blunt radius of a thicker wing but I could be wrong.

How a wing behaves once the stall is under way is a much more complex situation and can even favour wing sections and shapes that actually promote a flow break away.

Sorry Simon, you have described the transition point that is when the laminar boundary layer flow separates (separation bubble) and then reattaches as a turbulent boundary layer.  Generally speaking if the flow remains laminar that is fine in a pressure gradient where pressure is reducing due to the flow speed increasing.  When a laminar boundary layer encounters the opposite, a slow down in flow speed and an increase in pressure gradient then it breaks away completely but even that is not a stall.  It merely reflects the fact that at a non-dimensional number called the Reynolds Number that is on the low side flow will be laminar whereas above a certain number the flow transitions from laminar to turbulent boundary layer and the turbulent boundary layer will cope with the adverse pressure gradient that causes the laminar to break away.   Small scale models operate in the laminar flow regime and so only the front part of the wing, usually up to the thickest part of the wing, generates lift.  The same goes for the tail plane which is why small scale aircraft tend to have larger than scale tail feathers otherwise they don't fly well at all!  Larger scale models cross into the turbulent boundary layer territory and all of their wing and tail plane work hence that is why big models fly far better than small ones.

 

A stall occurs when the wing is at an angle of attack when the boundary layer, laminar or turbulent, will not stay attached and the flow breaks away and that is called the stall.  It will always occur for a specific aerofoil at the same angle regardless of speed.  That's why there are high speed stalls where the aircraft is a very long way from its "stalling speed" and yet it will stall if the pilot increases the AoA to the stalling angle.  In full size aviation, a stall speed is arrived at by reducing airspeed at 1 knot per second to give the unaccelerated stalling speed.  A high speed stall can occur at any speed provided the elevator is powerful enough to increase the wing's AoA to the stalling AoA.  Again, in full size, the pilot is usually warned of this by buffet as the stalling AoA is approached and recovery merely requires a slight reduction the force they are pulling on the stick.  Hence the term in air combat of pulling to the buffet which is where you are getting the maximum turning performance of that aircraft.

 

As regards the OP's question then, it depends on the Reynolds number that you are operating at.  Reynolds number (Re) is the relationship between inertial and viscous forces - to give it it's formal designation.  It is calculated by dividing (air density, air speed, a typical length e.g wing chord) by (air viscosity).  It's generally a large number with the divide between laminar and turbulent being around an Re of about 500,000.  The wing section and smoothness (or roughness) of the wing surface also affects the answer.  As an example of this, a full size glass fibre glider had a stalling speed of 38 knots when the wing was dry.  If you flew through a rain shower, the rain droplets on the wing at the leading edge would cause flow breakaway in the laminar flow part of the wing and stalling speed went up to 50 knots!

 

Unfortunately, aerodynamics is very complex so we generally operate to what was done in the early days of aviation and learn what works, or doesn't work, the hard way and build up our approach accordingly.

[J D 8 - Moderator]

     Long answer to a short question there Peter and as good an explanation that most can understand given the complexity of the subject. 

[Peter Jenkins]

Thanks JD8.

[Mark Hewett]

Hi Peter, I'm interested in what you've written about Reynolds number and how it relates to our model aeroplanes. I've flown models from indoor eflite Vapor that flies at walking pace, to 1/4 scale, and full size from microlights to B737. I find that an aeroplane generally, model or full size, pretty much flies how you intuitively expected it would. But I have been surprised how differently my 1/5 scale Thunder Tiger J3 Cub flies compared to my 1/4 scale Hangar9 J3 Cub. Both are film covered ARTFs, four stroke powered, with a similar weight in proportion to their size, if you get my meaning. But they fly very differently. It's hard to describe, but the handling and low speed/stall characteristics of the 1/4 scale Cub don't just feel like a linear step up from the 1/5 scale, the 1/5 scale feels like a model aeroplane, and the 1/4 scale like a full size Cub. I'd guesstimate the stall speed of each model is about the same despite the size difference, and the 1/4 scale model cruises, approaches and lands at what looks like scale speed, the 1/5 model needs to fly faster than scale. The wing chord of the 1/5 model is 28cm and the 1/4 40cm. I guess what I'm wondering is, are they the sort of chord sizes between which Reynolds number predicts the change from laminar to turbulent boundary layer that you describe? 

[J D 8 - Moderator]

           One of the main issues with model planes and more so the smaller they are is that the air molecules they fly in remain the same size.

This is very obvious with model boats where the bow wave/wake of the water molecules give the game away that t is a model you are looking at. 

  Small models operate at Reynolds numbers that are pretty much off the scale of what should fly but we can get around by having them very light and/or have the power to fly fast.  Peter will likely have a better explanation.

[Peter Jenkins]

Hi Mark

 

I regret to say that JD8's is (sorry JD8) a complete red herring.  The size of air molecules is just so many orders of magnitude smaller than our every day world that having a model even 100 times smaller (that's 2 orders of magnitude or 10 squared) will not matter in the slightest.

 

Your issue seems to be whether Re (Reynolds Number) is the cause of the difference between the 2 different sizes of your Cub.  (BTW in my younger days I used to fly full size gliders and power and one of my favourite aircraft was the Piper Cub - this one with the unusual power of 135 hp.)

 

The answer is absolutely.  To determine the Re for both then you need to use the Re calculation of:

Air density x characteristic velocity x charcteristic length divided by air viscosity.

 

The figures used must be in the same measurement system of course.  If we use an airspeed of 40 mph or 17.9 m/s, and wing chord of .28 m we can plug that into the above calculation (using Air density at sea level and 15 C is 1.225 kg/cu mtr and air viscosity at 15C is 1.81 x 10 to -5 kg/(m.s).)

 

We get an Re of 3.39 x 10 to +5 or 339,000.

 

For the larger model at the same speed we get an Re of 4.85 x 10 to +5 or 485,000.

 

Generally, the cutover point is an Re.of 500,000 for the transition from laminar to turbulent flow although some use 800,000.

 

A B737 in the approach will be operating at an Re of 1.5 million and in the cruise at about 5 million.

 

I should point out this is a very simplified discussion.  Mass/inertia also changes the way in which our models feel when we fly them.  As you have found, Re has a profound effect on how our smaller models fly.  Large models fly better and are easier to fly (leaving gyros out of it for this discussion) than small models.

 

The problem we have is how the boundary layer works at different Re.  What happens in the boundary layer is the deceleration of free stream air to stationary on the actual wing surface.  In laminar flow, the speed gradient is high and the boundary layer thin although it does grow in depth the further along the wing you go.

 

A turbulent boundary layer has mixing between the layers and a much thicker boundary layer results.  This increases the drag of not just the wing but all parts of the airframe compared with the laminar flow condition.

 

As I mentioned earlier, a laminar flow will break away once it encounters an adverse pressure gradient i.e. the flow starts to slow down and the dynamic pressure starts to increase.  Generally, for a wing this will be at or close to its maximum thickness point often between 1/3 and 1/4 of the chord.  Hence my point that only a small percentage of the wing, or tail plane, works in the laminar flow regime.  That's why small models have to fly faster to generate the lift required to keep them in the air.  Also why their controls feel barely in control!

 

A turbulent boundary layer is not affected by the adverse pressure gradient and remains attached to the wing till the TE.

 

Of course, when the stalling angle of attack is reached then even the turbulent boundary layer breaks away causing the loss of lift.

 

I apologise for reducing a very complex situation to a relatively simple explanation and taking some liberties in the process. 

 

Edited September 5 by Peter Jenkins

[J D 8 - Moderator]

8 hours ago, Peter Jenkins said:

Hi Mark

 

I regret to say that JD8's is (sorry JD8) a complete red herring.  The size of air molecules is just so many orders of magnitude smaller than our every day world that having a model even 100 times smaller (that's 2 orders of magnitude or 10 squared) will not matter in the slightest.

 

 

  That's OK Peter just another one of those old modelers tales then.

[Peter Jenkins]

39 minutes ago, J D 8 - Moderator said:

  That's OK Peter just another one of those old modelers tales then.

I'm afraid so JD8.  I should bury that one good and deep!

[Cuban8]

Some excellent (and essential) bedtime reading for model flying insomniacs can be found in Martin Simmons's book 'Model Aircraft Aerodynamics' - a few around on the book seller websites for only a few quid. I have it in my library and it's worth getting a copy not just for sleepness nights 😀 but for other more digestable info for those of us who are not particularly enthused by flight theory beyond the basics.

All joking aside, it's a very complex subject that's difficult to grasp, I find. Pete's concise contributions here are very good.

 

I think it might have been in one of the model mags years ago that had an aerodynamics article by a leading light of the time who was discussing wing sections, and how changes in their profile etc affected their performance in all sorts of ways when designing one's next model.

I recall his closing remarks something along the lines of "you can take all the figures and L/D graphs on board - but in all likelihood, drawing around a favourite Wellington Boot sole will be easier and not that far off the mark"..............

Edited September 6 by Cuban8

[Simon Chaddock]

Perhaps this is why the true Clark Y section developed in 1922 is still remarkably successful particularly in model aviation.

 

[Peter Jenkins]

5 hours ago, Simon Chaddock said:

Perhaps this is why the true Clark Y section developed in 1922 is still remarkably successful particularly in model aviation.

 

"If it ain't broke don't fix" it is a good maxim unless you are searching for ultimate performance.

[Mark Hewett]

23 hours ago, Peter Jenkins said:

 

I apologise for reducing a very complex situation to a relatively simple explanation and taking some liberties in the process. 

 

 Thanks for the explanation Peter - and for pitching it at just the right level, so even a pilot can understand it😃 I had in the meantime watched a couple of YouTube videos on Reynolds number, and had a play with an online Re calculator, but only succeeded in confusing myself. But I feel I now have a basic understanding about something I've been curious about for a while. Thank you.

[Peter Jenkins]

Glad you found the explanation helpful Mark.  Don't bother with online Re calculators.  A good approximation is to use the air density divided by viscosity figures at 15 C and that gives a figure of 67,680 approx.

 

You need wing chord expressed in metres and speed in metres per second.  Then just multiply them together to give a close approximation to the Re you are trying to calculate.