Foiler Design

sigurd
the trade maintains the angle of attack constant for a given speed and lift.

The inverted "vee" foil was an idea for getting additional lift at low speed. I do not believe Ian Ward tried the idea. It was not proposed to flap the foil. Roll stability is dominated by damping generated by the sail and apparent wind velocity.The negative diahedral would be adverse but not significant.
alans

http://www.highspeedsailing.com

 

Thanks Tom. I think some more stuff is starting to sink in now.

Is it correct that the Di is proportional to L*tan(alpha downwash)? If not I will get frustrated.

Wasn't there something about a low AR having a greater max CL as well? Not really related to the pitch stiffness I guess, but I am wondering how come?

On another note: If flaps or AoA are changed quickly enough, can this introduce turbulence and extra drag? If so, is this something that should be adressed in the control system design? For instance introducing some mass somewhere in the linkage between wand and flap so that the max flap actuation speed is lowered?

Mikey, I read yesterday about an eppler generator, let me see if I can find back to it.

 

Thanks alans, so the V is in addition to the daggerboard foil and rudder foil?

I thought it was wardi who found that anhedral was too difficult to sail? Maybe it would be different since they are not the "main foils"?

Anyways, if stowing height would not be a problem, maybe they could be curved so that the underwater part was more horizontal - should be easier to balance in roll, then?

 

something like this perhaps? Or curved the other way, but then you'd have to have two.

 

Did any moths try a canard like this? a small planing unit (green) controls the AoA of the submerged wing, independent of boat pitch.
Could be designed to be stowed easily. Perhaps it, along with an extra large rudder foil, could replace the daggerboard foil. Or the rudder foil. Or, the daggerboard with foil could be retracted in strong winds? I wonder if it could be made to assume a high CL without stall when re-entering after an inadvertent jump?

 

hi mikey,

i don't know of a eppler foil generator, but there is a long list of them here inclusive coordinates.

good luck.
boogie

 

Canard

====================
Sigurd, Dr.Ian Ward has done a lot of experiments with a forward canard. I'm not sure if he's used the specific altitude control system you drew or not. I believe that in the early pages of this thread he talks about it and also under "Hydrofoils" on the Aussie Moth forum.
As you probably know, your drawing is almost an exact depiction of how the foil system on the Hobie Trifoiler works.
I'm very interested in a foil system that will provide efficient lift and low speed takeoff as well as have excelent re-entry characteristics when purposely jumped.....

 

doug, i suppose then that the trifoiler don't produce suck on the ww side?

i don't think the above system is so good for re-entering. but something that adjusts the freestream aoa of the canard as a function of canard depth is the ticket i think.

what you think of this? a not depicted sensor actuates the tail elevator on the little plane. sensor maybe could be sticking up from the plane itself so we dont have the bother with pulling the control linkage through the freely moving joint at the bottom of the strut. you would have to anyway though if you want to jump.

edit: if flappes were preferred they could be used and then the tail could be locked. then the neat flap-mounted sensor that dave Culp brought up in this thread could be used.

 

Sigurd
Ian Wards work was early in the piece when everything about foiling was difficult. Anhydral would be more difficult especially for the main foils. But today with many of the challenges solved it may be worth reconsidering anhydral.

However adding retractable foil area to enable early lift out is not necessarily a good solution.
Firstly the conditions for early lift out in minimal wind strength compared to high wind strengths are quite different. For minimal wind a high lift drag ratio is the aim while minimum drag is the aim of the later case. That is keeping the induced drag low is the aim for minimal wind conditions. To the first order lift to drag ratio is a function of span divided by the square root of total wetted surface area. L/D ratio is not a function of lift (AUW). So adding area doesnot necessarily lift you out early. Minimum drag will normally occur with a Cl in the 0.6 to 0.8 range and this determines the lifting area you need to lift a specific weight at a designated speed

alans

 

"So adding area doesnot necessarily lift you out early."

I agree, I was rather thinking about it in the way that reducing area at high speed and when floating could decrease drag. So you could choose the total area to be the optimal area for early takeoff, then when it picks up or there is too little for flying you could retract some.
How ever I think it might be difficult to beat the LD of the illet system at its most efficient condition (I think just a few knots above takeoff speed?). Possibly with a more divided area it would be hard to take off earlier than the fastacrafts - even though when taking off the fastacraft does not have an optimal span loading since both the elliptical main foil and the rudder foil are lifting.

what means "(AUW)"?

 

Hobie / jumps

============
Sigurd, the Hobie trifoiler does produce RM by the differential action of the dual altitude control systems-as does the Rave and SKAT.
My guess at this point in the development of my "jumping" system is that a wand controled
"conventional"(Bradfield) system(80% load on main foil, 20% on rudder foil-more or less) is the way to go. The system for jumping would have to have a manual overide of the wand system for jumping since a jump using a wand that operates normaly would result in the boat re-entering with the mainfoil flap at max NEGATIVE lift position(back end of the flap up) which would be likely to lead to a crash. Manual overide for jumping is fairly simple to engineer but still needs to be tested. Some work on the foil characteristics may be required as well. It is known,for instance, that the Rush Randle foilboard jumps and re-enters well but it uses a very inefficient foil shape(a delta planform)-inefficient for low windspeed takeoff.
Its also been shown that a model foil using the Bradfield wand system will jump and re-enter
w/o crashing-though by all that is holy it shouldn't as best as I can tell. It must have something to do with the instantaneous response of the wand as the foil re-enters.

 

The Eppler foils were designed using the Eppler code. It was probably the first airfoil design and analysis programs to be widely available, and has an inverse design capability similar in some ways to the MDES mode of XFOIL. It was the state-of-the-art in the 1980's and into the 1990's. The original source code was published in a NASA report (Eppler, R., Somers, D.,"A Computer Program for the Design and Analysis of Low Speed Airfoils", NASA TM-80210, 1980.), and updated versions are still available from Mark Maughmer in the US, and Prof Eppler outside the US. Today, XFOIL does a better job of predicting section characteristics and is the state-of-the-art for low-speed airfoil design.

The coordinates (and the Eppler code input decks) for the Eppler airfoils are found in the book "Airfoil Design and Data" by Richard Eppler, Springer-Verlag (Berlin), 1990. The book is out of print, but you may be able to find it sold on the net, or in the engineering library of a local university. The book is an excellent description of the modern approach to designing sections, although it is heavily tied to the particulars of the Eppler code. However, you can execute the design philosophies in XFOIL and learn a lot about section design.

I don't believe there is any symmetrical section that is 6% thick in the book. The E297 at 11.4% is about as close as he gets.

For thin sections like the NACA 0006, the section shape doesn't have a lot of influence. It's going to act pretty much like a flat plate, with a large leading edge suction peak, at any substantial angle of attack.

Paul Bogataj published some modern thin sections designed for yacht keels using XFOIL:

Bogataj, Paul, "Fast Shapes for Thinner Foils", by Paul Bogataj", Sailing World, October 2000, p.46-48.

The Bogataj sections would be your best bet.

 

That's exactly what it is!

Low AR has a higher maximum lift because of vortex lift. The strong trailing vortices curling up around the ends (or a highly swept leading edge) have low pressures in their cores, and this creates low pressure on the upper surface, and thus lift. But it also comes at the cost of tremendous drag. The essence of sailing is to get a high lift/drag ratio in all things, so this vortex lift is not very useful for high-performance craft.

When the lift is varied, a starting vortex is shed into the flow. If you do this a lot, it's a signficant source of drag. Just waggle your rudder and see both the vortices (if the rudder is transom mounted) and the effect on performance. Notice I said when the lift is varied, not when the flap is moved - the two aren't necessarily the same thing.

Putting rate limits in a feedback control system is generally a bad thing. You would like corrections to happen quickly. A small correction immediately is much better than a large correction later. Rate limits also cause phase lag, which can be absolutely devasating to control system stability.

Actually, the ideal control system maintains constant lift. The craft would fly level with no vertical acceleration at all. The control system would be working like crazy to compensate for the waves and other disturbances that would otherwise create a change in lift.

I've often wondered if it might be useful to add a bobweight to a hydrofoil control system to provide acceleration feedback in addition to the height feedback of the wand. However, it would probably decrease the damping unless rate feedback were available, too. I would make the bobweight contribution additive to the wand input instead of putting the two in series.

But what you really need is some way to generate integral compensation with a mechanical control system - which easy to do electronically, but not so easy mechanically. Integral compensation would enable better steady-state height control, and allow the integration of acceleration to get rate feedback for better damping.

A first order lag applied to the acceleration feedback (bobweight) would act like an integrator for motions quicker than the lag time constant, so the combination may help with damping rapid motions, too. Some sort of viscous damper applied to the bobweight linkage might do the trick. Getting enough control power out of such a system to move the flap while still maintaining the desired frequencies and damping ratios might make for a heavy chunk of gear.

 

That is known as a "Shutt strut" and was invented by Sid Shutt. It's been used in many human-powered craft ("Flying Fish" among them, IIRC), but I don't believe it's been used by a Moth. There may be issues with Moth class rules.

 

Adding area would allow you to lift out at a lower speed. But having the lowest takeoff speed is not necessarily the object.

A better objective is to fly in as low a wind speed as possible so as to maximize the amount of flying time you can get out of marginal conditions.

If you go with truly huge foils in an effort to take off at a low speed, it still takes a lot of breeze to get you to the low takeoff speed with all that claptrap dragging through the water. If you reduce the size of the foils, the takeoff speed will increase, but it won't take as much wind to get you there. If the foils are very tiny, the takeoff speed is high and, again, it takes a lot of wind to get to takeoff. In between these two extremes of very large and very small foils is the size that will allow you to take off with the least amount of wind. It's a Goldilocks problem - not too much area, not too little area, just the right area.

Here's another way to look at it. Induced drag is inversely proportional to the speed squared. So if it were possible to fly at near zero speed, the drag to do so would be near infinite. The drag from floating goes to zero at zero speed. But induced drag goes to zero as the speed becomes very high, which makes flying attractive. So somewhere between very low speed and very high speed there is a crossover speed. Below crossover, it's better to float. Above crossover, it's better to fly.

If the other drag sources than induced drag (such as skin friction from wetted area) go up with speed squared, then the minimum drag occurs at the speed where induced drag makes up half the total drag. But even this may not be the optimum takeoff speed, because the thrust available to a sailing craft depends on speed, too. So you really need to optimize the whole craft.

Greg Ketterman told me it's much better to err on the small side than to err on the large side when it comes to sizing the foils. Since I've never actually built a hydrofoil and he designed one that set a world's speed record, I think that's pretty strong advice.

All Up Weight, also known as gross weight.