Foiler Design

[FarmerColin]

Originally Posted by Doug Lord

The Rave multifoiler uses a wand whose pivot is near the back end of the main foils and close to midship. I've had a few hours experience on that boat in varied conditions and some friends have had hundreds of hours sailing the boat.I've got several hundred hours of testing rc models utilizing midship Bradfield wands with not one crash/pitchpole in widely varying conditions. An interesting thing is that the most experienced Rave guys have told me that the boat almost never pitchpoles or flies out of a wave.
On the other hand Moths with bow wands seem to have trouble with certain specific conditions in waves.
If you draw a moth flying level in waves in a crest to crest half length equal to the distance between the bow and the daggerboard and compare a midship(board mounted) wand with a bow mounted wand it appears that for(at least) that specific condition the main foil on the boat with the bow mounted wand will enter the wave with the flap calling for maximum lift where the mainfoil on the boat with the midship wand will enter the wave with the wand calling for minimum lift. Thus the boat with the midship wand would be far less likely to burst thru the wave and wind up crashing.
Considering an upwind speed of 13 knots( 21.9 ft.per sec./6.68m/sec.) and a downwind speed of
22 knots( 37 ft.sec./ 11.3 m/sec) it seems to me with "normal" wand response that the boat with the bow wand would be far more likely to come out of the wave into air on the other side than would the boat with the midship wand.
Any comments?
Have any of the Mothies that have designed their own boats experimented with a midship wand?

Doug I'm not convinced that the midship wand is less prone to the problem you mention. One can draw similar diagrams - say with the wave crest at the midships point causing the foil to command maximum lift with the wave trough at the bow where the midship wand is being fooled into making what appears to be the wrong decision. However, one thing that can be said for the fore/aft location if the wand(s) is the effect this has on the sensitivity of the wand to the attitude of the boat. If the wands were located so far aft that they were alongside the rear foil then the wands would be sensitive to height but substantially insensitive to the attitude of the boat. Conversely were the wand to be cantilevered a long way ahead of the bow (practically difficult I'll admit!) it would become extremely sensitive to the attitude. So moving the wand further forward will contribute to pitch damping. Of course the foils have an inherent pitch damping as a rising foil is at a lower effective angle of attack (at 7 m/s boat speed a 0.5m/s rate of rise will reduce the angle of attack by about 4 degrees). Perhaps the ideal fore/aft location needs to be calculated taking account of the moment of inertia in pitch, area and aspect ratio of the foils, fore/aft separation of main and rear foils etc. However, I suspect the optimum would only be right for one speed as the 'loop gain' of the system would increase as the square of the boat speed.

One advantage of having the wands almost co-located with the foils is that the wand will unload the foil as a wave trough passes by.

I guess in an ideal world one would scan the surface ahead of the craft and analyse this to minimise the chances of a foil hitting the surface or the hull kissing the surface but I think the required sensors etc are rather out of place in anything but the most exotic machine.

FC, welcome to the forum! Great post! From what I've observed it seems to me that boats like the Moth and the Rave to a lesser extent seem to pivot in pitch(to the extent they do ) around the main foil or just aft of there. I wonder if that doesn't cause some degree of pitch coupling with the wand resulting in some of the unpredictable behaviour we read about-particularly in Moths? What do you think?

[FarmerColin]

Hi Doug - thanks for the welcome! My normal stomping ground is with aeroplanes and helicopters so I'm having a great deal of fun trying to understand how much of the ideas from those areas can translate across to hydrofoils. It occurs to me that hydrofoils seem to do remarkably well will the wand 'feedback' which effectively links the lift of the foil to a displacement. In most control loops (I'm thinking particularly of the tail rotor gyro systems on helicopters) it is necessary to feed into the equation displacement, speed and acceleration terms in order to get accurate and stable control. My guess is that the natural damping from the rise and fall of the foil is key to why we don't need a more complex control system. (by contrast the tail of a helicopter has very little natural damping and is 'fun' to stabilise).

I read something a while back by Shuttleworth about the Paul Ricard trimaran and he was ascribing its resistance to pitching to the very forward location of the side foils. Certainly the further the distance between the front and rear foils (what would be called the tail moment in aeroplane) the greater the natural pitch damping. However, if the fixed wing stability requirement is applied its important to load the front foils more than the rear foil (so that nose-up pitching increases the lift of the rear foil by a bigger factor than the front foil and thus shifting the centre of lift rearwards in opposition to the original attitude change. So in shifting the front foils forward its important to not shift too much load to the rear foil.

As you can imagine I'm having trouble juggling this around. I'm wondering if I should write a simple numerical model into which I could push some of this stuff and try to get a handle on the size of a few of the effects.

Fc, I had the great pleasure to work for a summer with Dr. Sam Bradfield, designer of the Rave ,Skat and some record breaking foilers. His help was instrumental in my being able to design my F3 RC foiler. His basic system on a Rave type foiler was to use 80% of the load on the forward foils with an angle of incidence of +2.5 degrees on the forward foils,0 degrees on the rudder foil referenced to the static waterline. Worked perfectly on the F3 and on my 16 footer(there were other problems there ,though).
Thought that might give you a known starting point.
By the way,in the mid 80's I designed a video piloting system for rc helicopters that allowed you to fly watching a tv screen. Among the most fun I've ever had using two hands!

[FarmerColin]

Doug - many thanks that should give me a good starting point. I'm thinking of writing a model that just looks at three T foils with defined area and aspect ratio and control surface chord (assumed to be full span) - I thought it would be interesting to consider the possibility of a wand control on the rear foil as an option. Be able to define the dimensions of the triangle they form and to be able to set a weight, CofG location, and roll and pitch radii of gyration. Then I think we need to be able to define the wand locations and the wand to flap gearing. Oh, and I guess we need sail centre of effort height etc and some way of defining the sail forces in the simplest way - perhaps just lateral and longitudinal values.

My thought is that initially it would be useful to try the model with a flat surface but with the system displaced in various ways and look at the transient response and then work on trying to input some realistic sea state info and sail force fluctuations.

(BTW I got into RC helis in the early '90s and gave up the day job to work on them full time in '95 - best move I ever made!)

Check out this extraordinary foiler animation:
http://members.aol.com/HYDROSAIL/

Scroll down to"Hydrofoil Control System Animation"
Illustrates Dr. Bradfields planing wand very well.

[sigurd]

Tspeer said: "These sections were an exercise to see if blunt-edged sections could be competitive with sharp-edged sections. The CFD results say, "yes," but I've heard anecdotal evidence that they may not work nearly as well as predicted. But I don't have any empirical data to support that."

I heard of P31202 (IIRC) used as rudders in a proa with not so good results.
If I interpret your xfoil graphs correctly, then the cambered P3 sections would be not so good near zero lift, so maybe that is important to avoid. In my sketches of diverse kite foiler board-like things I put a zero camber P3 in the horizontal wing because of the vertical lift of the kite.

[FarmerColin]

Sigurd - I think it is quite reasonable to have some loss in performance at low lift coefficients for sections with blunt trailing edges. For the section to produce lift it must establish a bound vortex around the section and to do this it must leave behind a starting vortex. An attempted flow around the sharp trailing edge (which requires a high acceleration in the flow at the trailing edge) promotes the shedding of the starting vortex. Indeed it is one of the boundary conditions in the theoretical treatment of sections that the flow around the trailing edge is zero. Without the sharp trailing edge it is possible to have flow around the trailing edge which will inhibit the establishment of the bound vortex resulting in little or no lift.

Doug - Many thanks. I was wondering about a couple of aspects of the wand operation. Since it seems to rely on the drag of the water to pull it backwards it presumably, at a given boat height, gets deflected back further the higher the boat speed. This would make the deflection of the flap be down further at higher speed causing somewhat more lift and raising the boat even further.

The other thing is the fact that, at the high riding end of the range the wand would appear to move through a greater angle for a given height change than it does with the boat lower down. This would seem to increase the 'gain' of the wand's authority when the boat is moving fast and where the gain of the system is naturally high. I wondered if any oscillations had been observed?

F, I'm not sure you realize that the wand has a shock cord pulling the wand "against" the water so as the boat rises the wand moves forward reducing flap angle or even going to up flap for negative lift. On the Rave and my F3 the two independent wands not only control altitude but develop all the righting moment for the boat as well.

[SimonN]

Originally Posted by Doug Lord

F, I'm not sure you realize that the wand has a shock cord pulling the wand "against" the water so as the boat rises the wand moves forward reducing flap angle or even going to up flap for negative lift.

On the Moths, setting this to the correct tension for the conditions is pretty important. Some of the guys even have the shock cord on a cleat, so they can tighten and losen it for the conditions.

[FarmerColin]

Hi Doug - Yes I realise there is a shock cord providing a restoring force. In fact that was one other source of a rising gain with height effect I was wondering about. Since as the boat rises the shock cord tension is falling this would also tend to make the wand angle Vs boat height more sensitive as the boat gets higher in the water. I guess the longer the shock cord the less the 'spring rate' of the cord will change with displacement.

[SimonN]

Originally Posted by FarmerColin

Hi Doug - Yes I realise there is a shock cord providing a restoring force. In fact that was one other source of a rising gain with height effect I was wondering about. Since as the boat rises the shock cord tension is falling this would also tend to make the wand angle Vs boat height more sensitive as the boat gets higher in the water. I guess the longer the shock cord the less the 'spring rate' of the cord will change with displacement.

This is certainly true with a Moth and the bow mounted wand. The elastic is about (unstretched) 1.2-1.5 metres long which gives it enough spring. Also, the range in which the elastic has to move is significantly less on a Moth than on the Rave system.

FC, you had mentioned doing a model simulation-if I understood you correctly. I'm very interested in seeing a comparison between a midship and bow wand on a monofoiler. I'd like to suggest that as an avenue of exploration for you for purely selfish reasons-I'd like to know which seems like it would be best as a guide on which to incorporate on my new foiler. I'm going to use a manual system with twist grip hiking stick control of the mainfoil flap with the rudder foil flap mixed in as well. I want to spend considerable time with each system to explore it thoroughly-they will be interchangeable.
There are advantages I can see for a midship or bow wand-and for racing and just fun -a manual system.
Interested in any thoughts you might have....

[FarmerColin]

Originally Posted by Doug Lord

FC, you had mentioned doing a model simulation-if I understood you correctly. I'm very interested in seeing a comparison between a midship and bow wand on a monofoiler. I'd like to suggest that as an avenue of exploration for you for purely selfish reasons-I'd like to know which seems like it would be best as a guide on which to incorporate on my new foiler. I'm going to use a manual system with twist grip hiking stick control of the mainfoil flap with the rudder foil flap mixed in as well. I want to spend considerable time with each system to explore it thoroughly-they will be interchangeable.
There are advantages I can see for a midship or bow wand-and for racing and just fun -a manual system.
Interested in any thoughts you might have....

Hi Doug - Yes, I think it will be fairly easy for me to make the model so we can try any arrangement of wand positions and foils positions. I have been wondering for example about a foiler catamaran with the foils supported on the cross beams so that they could be lifted completely clear of the water between the hulls. I'm going to write it in Delphi as its an almost self commenting language that others should be able to tweek to add any features I haven't put in.

The manual or semi manual system interests me. I like the idea of a joy stick in the cockpit so you can fly it in difficult conditions. Few automatic systems can beat the Mk1 eyeball!

[SimonN]

Originally Posted by FarmerColin

Hi Doug - Yes, I think it will be fairly easy for me to make the model so we can try any arrangement of wand positions and foils positions. I have been wondering for example about a foiler catamaran with the foils supported on the cross beams so that they could be lifted completely clear of the water between the hulls. I'm going to write it in Delphi as its an almost self commenting language that others should be able to tweek to add any features I haven't put in.

The manual or semi manual system interests me. I like the idea of a joy stick in the cockpit so you can fly it in difficult conditions. Few automatic systems can beat the Mk1 eyeball!

The main debate that has been going on is about bi-foiling monohulled boats. Tri-foilers are, imo, very different animals. Let me explain how we currently sail these bi-foiling monohulls. There are 3 controls on the foils. First, there is the wand, mounted on the bow. Next, there is the rudder which has either a flap that is manually controlled, or the whole foil moves and again, that is done manually. Finally, there is human setting of the angle of the main foil, which is done by the shifting of body weight. At low speeds, when we want go get up on foils, we sit back, increasing the angle of the foils and gaining more lift. Once up on foils, the wand and flap on the main foil aren't enough to stop the boat continuing to climb until the foil comes out. You have to move forward, to reduce the angle of the foils. So, the helm is constantly making sure they are positioned in the boat the present the optimum angle of the foils.

You can also effect the angle of the foil by the lift from the rudder. This can be used to set the angle of the whole set up. This is important as you want a certain balance between the 2 foils. If you put too much on the front foil, yoiu are risking a big wipe out.

Where this gets fun is when there are waves. We know that hte quality of the control systems make a big difference as some boats are almost unsailable in waves while others are relatively easy. Downwind is far harder than upwind.

Now, the real crux. While manual control might seem great, they have serious limitations in race boats. When racing, you want to be able to spend as much time as possible with your head "out of the boat". You need to be looking at where you are on the course, where all the other boats are, when to tack/gybe, what the wind is doing, tides etc. The more manual the system, the less you can look around. The penalty for getting a wave wrong is very high. You therefore need to consider whether the amount you can improve by having manual control will overcome the inevitable problems of either catching waves or not getting your head out of the boat. If you are building a boat just for straight line speed, maybe manual controls would be best, but for racing, well, non of the Moth sailors believe that it will be as quick. Doug believes that we would learn how to sail the boat manually and not miss waves, but what happens when a boat you didn't see calls "starboard" at the last second and you are concentrating on avoidance just when bad patch of waves comes through.

Hope that helps!