Kite Sailing

A different approach derived from contemplating what characteristics kites will have to have in order to be practical in the real world of yachting and large boat hauling rather than by extrapolating from my experiences and hopes as a committed kiteflier as I've generally done up to now- not being alone in this either!   .

Since my naive dreaming of kitepower in the '80's (when I rushed out and bought a 12m  'Formula 40' high performance catamaran with the intention of sticking a kite on it then and there) , I've basically  been holding back from trying to design or build kites for large boat kite sailing because I didn't believe it could be practical (or safe) until we have kites with 90% or so de-power- apart from for low performance kites that can only provide downwind course options.

Now (2005+) we DO have controllable 90% de-power kites (Bow type LEI's for e.g.)- or at least we will have within the next year or so- when steering problems in de-power mode are solved- and more than one type for sure.

But now I don't think this is enough.

As yet, sailors and seamen have no conception as to just how difficult it is to keep high performance kites flying.

If they think about it at all, they probably imagine that it will be like trimming a spinnaker.

Well I have news for them, it's not- it's orders of magnitude more difficult, even during daylight hours in moderate and steady winds.

It's so difficult as to limit the likely use of kites on larger boats to low performance (L/D less than 3), 'spinnaker replacement' styles only useable for downwind courses, or to 'adventure' sailing- for which  difficulty,  uncertainty,  and risk are reasons to do it rather than reasons not to.

Even in good mid range wind conditions (which I still hope to experience some day!) and in reasonable seas,  currently available kites need to be continually under the active control of skilled fliers- and even then, will crash or rip their rigging  apart too often to be acceptable for mainstream yacht or large boat use I believe.

Kite traction hopefuls like myself have been pinning our hopes on improved kites and on developing kite auto pilots that will take away  the hard work and uncertainty, making it possible for yachts to use kites with a degree of reliability and performance not less than with current sails- or not much less. 

And if this isn't true, what reasons do we have for thinking that kitepower will ever be useful for larger boats?!

Traction kites ARE getting better; wider wind ranges, more reliable and easier control and much more de-power.

But sober consideration suggests that our expectation of the development of useful auto pilots for kites may be misplaced.

Consider parallel fields for which auto-pilots have been developed:

The first thing that an aeroplane pilot does when faced with any out of the usual situation is to turn the auto pilot off and take personal control- and aircraft auto pilots have already benefited from billions of dollars and 50 + years of development- which admittedly we can now draw on for kite auto pilot development.

Except that yacht auto pilots have been developed by drawing on this prior art - and they don't even try to control the sails- all they do is steer the boat- because sail dynamics are too complex. 

Maybe we have to say this quietly for now- but kites operate in three dimensions not just two- their dynamics are massively more complex than sails on masts.

To me it seems that the flight envelope for kites is just too indeterminate for auto pilots to have much chance of being comprehensively useful. 

Kites stall, they overfly, their lines are slack sometimes for extended periods, they are subject to totally unpredictable winds that shift 90 degrees in seconds, and the apparent wind they are subject to during any normal flying cycle ranges from negative to four or more times the true wind speed. 

Of  the inputs that are potentially available for a kite auto pilot, by far the most powerful set; line angle,  direction and tension at the tether point,  are all corrupted by wave action.- while knowing the wind direction and speed at the boat is not of much help because, annoyingly,  kites are in constant motion, up, down and sideways relative to their attachment point.

But if other inputs (like inertial guidance and horizon detectors for example) can substitute, if it's true that even experienced fliers will fail to keep kites flying more often than is going to be acceptable for mainstream yacht or large boat use then isn't it too much to expect that any electro- mechanical system can succeed?

It may be that the most that kite auto pilots will be able to accomplish in the foreseeable future is to offer some assistance to a full time, fully focused, experienced flier in smooth wind and for a narrow range of flying circumstances- if that.

And even to accomplish this limited goal, there will need to be many years of development and very considerable investment, - and then there is still the problem of supplying power at the kite for the continual control system actuation that auto pilot use will demand.

Misery!

But there could be an answer- and it comes from the early days of heavier than air flight.

The Wright brothers believed that aeroplanes should be inherently unstable- as traction kites are today.

They believed that only by deliberately building in instability would controlled flight become possible.

They were spectacularly wrong in this.

Within 5 years of their first flight, other experimenters had capitalised on this error and designed aeroplanes that were inherently stable- that looked after themselves unless the pilot chose to input some directional change.  Within 10 years of their first flight, the Wright's design principles, and their aeroplanes had been consigned to history. That I can think of, nothing significant they developed has become part of the modern aeroplane- though their position in history is secured by one thing they did right; showed that heavier than air flying is practical.

100 years later, almost all aircraft are designed to be inherently stable.

We need to adopt this approach also if kites are to become useful for larger boats.

Except in athletic show-off sports like buggying, kitesurfing and kiteskiing, traction kites should be auto stable- that is, their default flying mode should be flying at apex directly downwind- and fully de-powered.

Or at least they need to behave this way if kite use with larger boats is to progress past being a succession of over-hyped money wasting failures that leave bad memories enough to further inhibit the development of practical kite traction for years into the future.

If performance traction kites could be developed with this inherent stability characteristic, they would make trans-oceanic kitesailing safe and practical.

They would also make any limited forms of kite auto pilot that may be possible much more effective:- because it will be easier to do for kites that are already inherently stable, and because when any situation arose that such an auto pilot could no longer deal with, it could be programmed to turn itself off while the kite gets itself back into a safe de-powered stable position. 

But what is this magical thing; a kite entirely without any external or stored power source and no dedicated control surfaces or means to actuate them, that gets stability by its shape alone, even to the extent that it recovers by itself from serious turbulence and wind shifts?  Could such a thing be possible?

It is, it's the single line kite: and has been around already for many thousands of years and is now known in a huge variety of forms, albeit generally with aerodynamic efficiencies (that is, lift to drag ratios; L/D's) insufficient for performance traction kiting. 

Is it possible to build a high performance traction kite that behaves as a single line kite when left to itself?

This is indeed where things get tricky.  Single line kite designing is a black art indeed- MUCH more difficult than traction kite designing- and this is true though it may come as a big surprise to many traction kite designers.

Traction kites are built according to well known and understood principles - albeit that the task becomes more difficult the closer to various performance edges a particular design is pushed to.

By contrast, the stability of single line kites is an interplay between the discontinuous aerodynamic effects of turbulent flow, inertia, and the weight force - creating a complex feedback dynamic which is stable for only a few of it's almost infinite possible combinations.  That there is even such a thing as a self-stable single line kite continually amazes me.

There is no current body of theory by which single line kites can be designed, and there may never be.

Only one thing is known for sure; - that single line kites are pendulums, their upward seeking tendency derives from their centre of gravity being below their centre of lift. 

Unfortunately, as L/D (measure by angle of flight) increases, the ability of this pendulum effect to continue to point the kite in the desired direction diminishes. 

Fortunately, in practice, some known single line kites do reach the level of aerodynamic efficiency (say L/D>5) necessary for traction kites, and as there are some types now, it seems likely that others can be discovered or developed, which implies that the answer to the question above is, tentatively, yes.

I'm proposing a principle; that to be useful and practical, traction kites for ocean going use would be advantaged by having the characteristics of self stable single line kites- with steering and pull control then applied as and when required. 

That this might also be possible in practice is suggested by their being extant single line kites that have aerodynamic efficiencies commensurate with current high performance traction kites.

To test this further, for the last month I've been attempting to make an Arc style kite perform adequately as a single line kite.

This is not to imply that this style is necessarily the best form for such a kite- but they're convenient for me to work with and have the added advantage of exhibiting some single line type stability already.

If I can get an acceptable level of self stability, the plan will then be to use radio control for steering, power control, and to operate safety releases.

Wingtip drag flaps (courtesy of Bas Lansdorp at Delft University) should be ideal for steering.  These have the huge virtue of working perfectly in stall and in power-up mode while still working proportionally in full de-power; when the kite's angle of attack is so low that "normal" type kite steering becomes mushy then reverses.

Most usefully of all, I think it will be possible to make these drag flaps work within acceptable wattage limits and to scale them to kite sizes of 300sq.m or more.

For power control, I'm thinking of using winches controlling a series of spanwise cords- equivalent to A (leading edge), B, C, and D (trailing edge) bridle positions on a 'foil kite.  I've tested this system in principle by fitting ground adjustable cords- and they do grant wide ranging control of profile, angle of attack and hence of pull. 

These winches will be big power users though- relative to steering- but my belief is that for large ocean-going kites, power control can be exercised quite rarely, and slowly- over minutes rather than seconds. I'm also sure that an over-riding de-power system can be devised that will let things off in a big hurry when necessary - and using very little wattage. 

There can also be a line release on each wingtip, by separate channels- release one or, in extremis, both.

At this stage the criteria that have to be met before it will be useful to move on to steering and power control are;

Minimum wind for reliable hands-off stable flying: 1m/sec to 2m/sec.  At less wind than this current traction kites won't function anyway.

Maximum wind while retaining single line stability 10m/sec-15m/sec. Enough for now.

Ability to self recover without flier input from the edge, from slack line drift backs, from side winds, and from turbulent incidents.

L/D greater than 5.

I'm not at present considering ease of launching as a criterion- because there are now so many launching options that one or more is sure to be useable - and because such a self stable traction kite would not require frequent relaunching anyway, so launching can be comparatively tedious provided it's also reliable and safe.

The starting point was a standard 20sq.m kitesurfing style Arc (approx. 12 sq.m projected) and for comparison I fly one of our standard 8sq.m single line Pilot kites at the same time - because these Pilot kites meet all the above criteria other than L/D.  They are known as very reliable fliers and are in widespread use.

It's true that all Arcs auto apex to some extent, but they don't de-power enough for one-size-fits-all (well nearly all) trans oceanic sailing, tend to hang to one side or the other, do not recover reliably enough from the edges or drift backs, and require far more wind speed to stay aloft than do dedicated single line kites.  This last is important - traction kites like Arcs can only be kept aloft in very light winds by flying a continual figure eight pattern to generate higher apparent wind speeds.  Single line kites don't have this possibility.

To gain the required lower apparent wind threshold, I've progressively changed the prototype kite's form (by what we call wedging) and it will now fly by itself in very light wind- similar to the lightest an 8m Pilot will stay up in, and within the proposed 1m/sec- 2m/sec range I think.

At the same time, I've had to make many other minor form and profile changes in order to keep it's stability and L/D at the required level.

It's upper wind speed stability limit when flying as a single line kite also now meets the required goal, and as an expected bonus from these changes, it's apexing is now more central and reliable than for standard Arcs, and it's pull (because it's angle of attack is now nearly zero) is substantially reduced,- which makes it likely that the required range of pull control will be available.

As yet it's ability to recover from the edges, from drift back and from turbulence is not as good as for the Pilot kite flown for comparison. This testing has been done in bad wind at inland sites though, so maybe it's good enough but I'm not yet convinced.

 Based on the rapid progress to date, I'm hopeful of getting further improvements.  If it's capacity to self-recover improves just a bit more it will definitely show that the principle I'm proposing is practical.

Currently I launch SLARC1  by it's  V lines of 15m length, starting at the tips when space is limited, then let it out on a single line to the chosen altitude. Once up, it flies reliably as a single line kite in a wide range of winds, hands off hour after hour at any altitude you choose- and at an exceptionally high angle of flight.  

I've calculated that the line drag of the 20 sq.m SLARC 1 (actually it's number 3, the series started 3 years ago, but I've started the clock again) when 65m's out is the same as it used to be on conventional four lines at 25m.  This is based on each of the original four lines having been required to accept up to half the kite's maximum pull and assumes that drag is proportional to diameter times length- no Re considerations.

For traction kiting this will be a considerable advantage.

If it does prove possible to get an acceptable level of self recovery from this style of kite when flying single line and I'm able to move on to fitting steering and power control,  I see no disadvantage upwind relative to any other traction kite- in fact because the SLARC is fully functional down to lower angles of attack than most (all?) other  traction kites, it's L/D can be expected to be better- and certainly looks to be.

Nor do I see any disadvantage while reaching- sure, 'conventionally' controlled kites can be 'worked' more- for a while- but right around the world?- I don't think so.

For downwind and on broad reaching I see major advantages- what conventional sail or conventional kite powered craft could hope to match one hauled by a continually looping L/D 6 kite out at 200m's or more?  

And, in the long reaches of the night, when the crew have better things to do, this kite will just hang up there looking after itself- while there's enough wind that is.

The winch system required for kite control would be considerably simplified by having to deal with only one line; the ideal of flying on a short line upwind but out long for downwind will be easily accomplished.  Kite looping will not twist the lines as there's only one and it can be swivel attached at the V. Letting out as a gust response will become an option, as will pulling in to bridge wind lulls.   Best of all, with only a single line to deal with, attachment systems to the boat can be arranged so as to almost eliminate heeling caused by kite pull, which will certainly make monohull configurations- the fastest and most efficient possible layout- practical for larger size kite boats.

Nor is it only large boats that can benefit from this approach to kite traction.

For buggying, old codgers like Scott (Skinner) and I will soon be content to cruise around the desert under kites that look after themselves while we have the occasional senior moments- but still have enough performance to give everything else a fright over longer courses. 

For Antarctic journeying- who's scared of crevasses when the kite you're hooked under will sit up there awaiting a command to haul you back out?!

And consider the kitesurfing and snow kiting jumps that will be possible when a de-powered high-performance kite is suddenly turned skywards and powered up from a high speed pass at 100m's or more out.

And to add another level to this approach; a small auto stable kite, rigged to and flying above a larger  traction kite that doesn't have this self stability will provide the same level of self recovery and hands-off reliability as would be possible from a single larger auto stable kite.

Very large kites (for very large boats!) will not themselves have to be auto stable, but could be controlled by small (say 25% of the main kite's size or less) auto stable steerable kites flying above them.  - and not least to assist launching.  Provided they had good gust response (that is, some degree of automatic de-power) like delta style framed kites have for example, it may not even be necessary for this top kite to have any volitional  de-power  at all- which simplifies things a lot.

The first step is to get a high performance traction kite to function as a reliable and dependable single line kite.   SLARC1 is 90% there now, but experience tells us all that the last 10% is the most difficult bit. 

Even if the Arc based approach is not successful, other styles of kite still might be.

If such a kite, of any scaleable type, can be developed, then the next steps will be to fit steering, fit power control, devise launching systems and scale it up.  All of these things seem to me to be possible, it's the first step that's the key. 

If successful, it truly could make large boat kitesailing practical - and not just for crazy people like ourselves or for downwind courses or for 'adventure' sailing.

A direction worth pursuing!

                                              Peter Lynn, Ashburton NZ Feb 27, 2006