sail aerodynamics

[jam007]

Originally Posted by yokebutt

Anders,

Is that also true of the same section without camber? Or is that why cambered foils have their low-drag bucket shifted? Please explain!

It is no clearcut answer so here goes...

Some important terms:
Angle of Attack: The angle between the airflow and the mean cord of the foil.
Lift: The force perpendicular to the flow of air
Drag: The force in the same direction of the airflow.
Cl: lift coefficient. A non dimensional number describing how much a lift a foil creates. To get the actual lift force you use:
Lift= cl*density*area*(velocity^2)/2
Cd: drag coefficient. same as lift coefficient but describes the drag.
Stall: When the flow no longer can follow the foil smootly. Instead a wake of turbulent air is created. (Like the suction area behind a lorry) At a certain angle of attack any foil starts to stall. When a foil stalls the lift decreases and the drag increases.

If there was no drag all foils would create a net force perpendicular to the airflow. The drag shifts the net force so it is angled slightly along the air flow
the angle can be calculated by:
drag angle=arctan(Cd/Cl)
See picture in my previous message
The Cl/Cd ratio are the non-dimensional version of the Lift/Dag ratio.

If a foil creates a forward net force as seen from the coordinate system of the foil depends on both angle of attack and dragangle. If drag angle is less than the angle of attack then the net force is forward.

For the NACA 0012 (The uncambered version of NACA 6412) the drag angle is 5 degrees (arctan(0.074/0.87)) at an angle of attack of 10 degrees. This means that this symetrical foil have a net force directed forward.

As understood from the above (hopefully) Cl and Cd and therefore the drag angle are not fixed numbers for a foil but depends on angle of attack. For every foil there is a certain angle of attack that gives the lowest drag angle.

A cambered foil is simply put a bent symetrical foil. A symetrical foil has its lowest drag (and lift=0) at 0 angle of attack. A cambered foil has a leading edge that will be straight towards the oncoming flow at a angle of attack different from 0 and have low drag around this angle. This shifts the bucket and gives a higher cl/cd ratio and therefore a lower minimum drag angle.

But remember that the foils I have shown are 2D foils. Real 3D foils have higher drag than their 2D counterpart (due to induced drag but thats another story). For a 2D foil the Cl/Cd ratio can easily be 50 or more, a 3D wing around 20, but a sail have a Lift/Drag ratio of around 6. A Lift/Drag ratio of 6 means a drag angle of 9 degrees. This drag angle is so high that the net force is more or less perpendicular to the mean cord of the sail.

Sails have a maximum Lift/Drag ratio at an angle of attack that is lower than the angle of attack of maximum lift, just before stall. Going to windward the lift/drag angle is important since it can be shown that this ratio directly affects maximum velocity made good.

C A Marchaj mentions in "Sail Performance" the interesting effect of pumping with the sheet to create "negative drag" by using unsteady flow effects.

I hope this gave some explanation. (And that it is in its essenssials correct. I´m not an aerodynamics expert just a physicist interested in sailing.)

Anders M

PS Had to check that normal is an english term and not a swedish after your reply. Hope I avoided any more strange terms, its so easy to write swenglish.

[yokebutt]

Thanks Anders,

Bethwaite had an interesting example on the effect of pumping. The comparison was between two boats sailing along in 5 knots of wind, one is pumping and the other one isn't. In this idealized example the boat being pumped has a wind speed over the sail of 10 knots half the time, and zero the other half. And of course, according our old friend, the rather unpleasant Englishman, the pumped boat is pushed along faster.

Another source to peruse if you are interested in unsteady flows is Steven Vogel's book Life in moving fluids. Very good and also very funny, one of my favorite statements of his is to the effect that "many people have a fuzzy notion that the boundary layer is a discrete region, rather than a discrete notion that it is a fuzzy region"

Yoke.

[mojounwin]

I've been brought up to believe the headsail provides roughly 80% of the drive and the main was more for providing balance, but why is it that modern sportsboats seem to be going with smaller headsails and bigger mains?

Cheers
Mike

[cyclops]

They like the old, Clippers, Tall Ships and the Wind Jammers, more than PR releases?

[Skippy]

I think if you look at similar multi-foil airplane wings, the forward/upper foil is usually small. Aerodynamic support from the main is one of the things that make the jib so effective, so there has to be some kind of balance. Other than the disappearance of older racing rules that used to encourage more headsail, the large jib will have trouble from sagging in heavy weather. And for whatever it's worth, there's a stability issue with the headsail(s) grabbing more air rather than less as the boat starts to heel. Also, maybe that combined with the longer foot could result in a genoa getting dunked more often.

[LP]

Here is a thought to ponder. Aerodynamic pressure on an airfoil is generally centered about 25% chord. A complex wing that has lift enhanceing devices, flaps and slats, is not considered as three separate foils, but as a single unit, even though in most respects the high lift devices have separated from the wing. The new chord length is now measured from the new position of the leading and trailing edges, plus the added camber as a result of the downward deflection of said devices.

Is there a more direct correlation between wings and sails with this analogy rather than bi-plane theory. If the jib/main assembly is taken as a single unit with total lift placed at 25% chord, the center of lift will fall squarely on the jib. Hmmmmm.

Another ponderance. I think it was earlier in this thread that a discussion on fractional vs. mast head rigs took place. The common statement was that on similar boats, the fractional rig was just as effective as the masthead. On early Learjets, they had a problem with flow separation as airflow over the wing approached transonic speeds. Granted, we haven't quite acheived these speeds with sailing craft, but bear with me. To continue, this flow separation occured in the region of the ailerons rendering them ineffective. Look Ma, I'm going supersonic and I've lost control of my aircraft. Learjet's "fix" was to install votex generators on critical areas of the wing. These were "T" extrusions about an inch in length with the top of the "T" bonded to the upper surface of the wing with a slight misalignment to the airflow over the wing. Thus, creating a vortex, energizing the boundary layer and reestablishing boundary layer flow. Now the ailerons could change the amount of lift the wing generated in that portion of the wing.

Now here comes the stretch(analogy). Does the lift vortex that is attached to the head of the jib energize the flow over the main in the vicinity of the jibhead. Providing the main doesn't impose an endplate effect, the vortex that is being shed at the jibhead will be more highly energized because the flow velocity in the vortex will be greater than the local flow. Thank's to Bernoulli this supplies an additional local pressure drop that adds to the differential between the the weather and lee sides of the main. If the main does provide endplate effect, then the above is hogwash, but now drag is diminished, due to endplate effect, as opposed to lift being enhanced.

And one more ponderance . . . maybe. Is there really such a thing as laminar flow on a sail? In regard to aircraft, laminar flow wings are extremely picky about surface roughness. Bug juice will trip the boundary and cause turbulent flow. Not to get confused, turbulent flow is still attached to the airfoil. It's just a little more draggy. I find it difficult to believe that, with the preponderance of obstacles that support a set of sails, there is any laminar flow anywhare on a set of sails. When the luff starts luffing and the tell-tails start telling(sorry, poetic license), personally I think we are seeing flow separation and not a transition to turbulent flow. Laminar flow over the sails is tripped way before it ever sees the sails.

There, you have it now. I've set myself up for a royal lambaste.

Regards to everyone on the site. It's a great place to hang out.

[Packeteer]

so when I build my wing sail I should look at aeroplane wings from bygone days?


ps. I haven't read the whole thread yet, my brain is mush after the first page

[tspeer]

Jimmy Dolittle III (test pilot & grandson of THE Jimmy Doolittle) calls vortex generators the "horns of ignorance". They are used as a quick fix for premature separation arising from all kinds of causes, like transonic shock waves (the Lear example), interference effects (look at all the VG's on the underside and side of a B-1's tail), bends in an engine inlet duct, and flap deflection. The vortices shed by the VG's entrain higher energy air outside the boundary layer and bring it down close to the surface, and they sweep low-energy air from the boundary layer up and away from the surface. Like the transition from laminar to turbulent flow in the boundary layer, this adds drag but makes the boundary layer more resistant to separation in an adverse pressure gradient.

VG's are generally small vanes with an aspect ratio less than 1, and can be rectangular, triangular, or gothic (curved) in planform. When arranged in a herringbone pattern, alternate vortices rotate in opposite direction. When arranged in parallel, the vortices are co-rotating. There are also ramps and triangular ramps, and Y-shaped VG's like the Wheeler Wishbone. Micro-vortex generators that don't poke up outside the boundary layer still produce measurable effects and a small (1 - 2 db) quieting of the flow noise. NASA has funded research on on-demand VG's that are actuated by MEMS devices. These are flush when they aren't needed, but extended when necessary. The synthetic jets used in many active flow control applications are basically pneumatic vortex generators.

The vortex shed at the head of the jib might help energise the boundary layer on the main, but it's much larger in scale than the boundary layer. Instead, hardware at the hounds, forestay, even the wakes of halyards are more likely to act as turbulence generators for the boundary layer on the mainsail.

I've not been able to find much in the way of a good picture of the jib vortex. This one perhaps comes closest.

You can just make out two "skeins" of twisted streamlines downstream of the leech, showing the vortex from the jib merging with the vortex from the head of the mainsail. If you look at how far down the lee side streamlines are lifting up, I think it illustrates how the "tip" vortices are not just shed from the tip, but are actually shed little-by-little along the span, with the greatest concentration being at the tip.

The vortex from the head of the jib has the effect of "lifting" the part of the mainsail above it and "heading" the mainsail below the vortex.

[SuperPiper]

[quote=Skippy]. . . And for whatever it's worth, there's a stability issue with the headsail(s) grabbing more air rather than less as the boat starts to heel. QUOTE]

This is a concept that I had never heard before. Can someone explain this for me?

[sharpii2]

Originally Posted by learpilot

Here is a thought to ponder. Aerodynamic pressure on an airfoil is generally centered about 25% chord. A complex wing that has lift enhanceing devices, flaps and slats, is not considered as three separate foils, but as a single unit, even though in most respects the high lift devices have separated from the wing. The new chord length is now measured from the new position of the leading and trailing edges, plus the added camber as a result of the downward deflection of said devices.

Is there a more direct correlation between wings and sails with this analogy rather than bi-plane theory. If the jib/main assembly is taken as a single unit with total lift placed at 25% chord, the center of lift will fall squarely on the jib. Hmmmmm.

Another ponderance. I think it was earlier in this thread that a discussion on fractional vs. mast head rigs took place. The common statement was that on similar boats, the fractional rig was just as effective as the masthead. On early Learjets, they had a problem with flow separation as airflow over the wing approached transonic speeds. Granted, we haven't quite acheived these speeds with sailing craft, but bear with me. To continue, this flow separation occured in the region of the ailerons rendering them ineffective. Look Ma, I'm going supersonic and I've lost control of my aircraft. Learjet's "fix" was to install votex generators on critical areas of the wing. These were "T" extrusions about an inch in length with the top of the "T" bonded to the upper surface of the wing with a slight misalignment to the airflow over the wing. Thus, creating a vortex, energizing the boundary layer and reestablishing boundary layer flow. Now the ailerons could change the amount of lift the wing generated in that portion of the wing.

Now here comes the stretch(analogy). Does the lift vortex that is attached to the head of the jib energize the flow over the main in the vicinity of the jibhead. Providing the main doesn't impose an endplate effect, the vortex that is being shed at the jibhead will be more highly energized because the flow velocity in the vortex will be greater than the local flow. Thank's to Bernoulli this supplies an additional local pressure drop that adds to the differential between the the weather and lee sides of the main. If the main does provide endplate effect, then the above is hogwash, but now drag is diminished, due to endplate effect, as opposed to lift being enhanced.

And one more ponderance . . . maybe. Is there really such a thing as laminar flow on a sail? In regard to aircraft, laminar flow wings are extremely picky about surface roughness. Bug juice will trip the boundary and cause turbulent flow. Not to get confused, turbulent flow is still attached to the airfoil. It's just a little more draggy. I find it difficult to believe that, with the preponderance of obstacles that support a set of sails, there is any laminar flow anywhare on a set of sails. When the luff starts luffing and the tell-tails start telling(sorry, poetic license), personally I think we are seeing flow separation and not a transition to turbulent flow. Laminar flow over the sails is tripped way before it ever sees the sails.

There, you have it now. I've set myself up for a royal lambaste.

Regards to everyone on the site. It's a great place to hang out.

Hi LP.

25% of the chord? Seems strange to me. Especially when that 25% is usually in front of the 'hump' of the airfoil shape. Also, if that were true, tailplanes on airplanes would be designed to pitch the nose down instead of the opposite, which seems to be, at least on the flying models I have had as a kid, common practice.

I was taught that the Center of Lift was aft the highest point on the airfoil section. And how far aft it was depended on several factors, not all of which where predictable. That's what all those expensive wind tunnels are for.

From what I understand (or have been led to believe), finding the center of effort on a sail rig is practically a black art.

My guess is that with a masthead sloop, the Center of Lift might be 25% of the mainsail's chord. I get this number by a very scientific method. It's called a barn yard guess. It is based on the idea that the jib is producing lift of its own as well as enhancing the lift of the mainsail.

But who knows?

Bob

[sharpii2]

Originally Posted by mojounwin

I've been brought up to believe the headsail provides roughly 80% of the drive and the main was more for providing balance, but why is it that modern sportsboats seem to be going with smaller headsails and bigger mains?

Cheers
Mike

Because proper staying geometry and hull stiffness are harder to obtain on todays lighter (DLs of well under 100 on some) boats with ever greater working sail plans. A big jib needs a relatively heavy (DL of around 200) hull to hold the luff straight. A jib with a saggy luff is just about useless for anything except ballancing the main.

I know, I have used mine for that very purpose. Half roller furled, it was about as aerodynamic as a toilet paper roll, but it did hold the bow off, so the main could do its job.

I also know from experience that a 3/4 fractional rig can be sailed with the main only. Mine rutinely was. I have never seen a masthead sloop do that.

Bob

[Skippy]

sharpii2: I was taught that the Center of Lift was aft the highest point on the airfoil section.

Is that relative to the foil's chord line or to the wind direction? I know "on the section" sounds like you mean the chordline, but most of the CFD pics I've seen appear to have a strong low-pressure peak just behind the mast or leading edge. That sounds more consistent with the other way -- highest relative to the wind, which looks to me like somewhere around 25% or so.


SP, just look at a jib that's sheeted out. It's canted INTO the wind. When the boat heels over a ways, the headsail stands upright.

[brian eiland]

Looking at a couple of the recent postings, I thought it might be a good time to re-introduce a reference site I made back at posting #67, Avrel Gentry's Updated Web Site

Or for a more illustrative view, have a look at Paul Bogataj's simple but effective article at http://onedesign.com/articles/article6-1.html . Notice in particular his location of the chords of the airfoils, particularly that of the mainsail.

[jam007]

A simple experiment rearding position of maximum lift. Drop a rectangular pice of paper, lets say 15 cm x 3 cm with one of the long sides pointing towards the ground. It will not fall straight down but at an angle and start to rotate backwards as it falls. This is because the centre of lift is forward of the papers centre line.

Anders M

[gggGuest]

Originally Posted by sharpii2

Because proper staying geometry and hull stiffness are harder to obtain on todays lighter (DLs of well under 100 on some) boats with ever greater working sail plans.

Even supposing that modern structural engineering and materials doesn't produce stiffer boats than wood or aluminium boats the engineering of thirty years ago (which of course it does) the fact that no open rule dinghy or catamaran ever even considers a masthead rig should be enough to point out the flaw in that theory. The masthead rigs were all to do with rating rule benefits and the easiest way to get more rag on a small spar.