Previous Page | Table of Contents | Next Page

Amateur Yacht Research Society, February 1997 Newsletter

And a further contribution from the AYRS internet list, a discussion started by Dave Culp;

Subject: Pumping: Legal? Fair? Advantageous?

From: Dave Culp Date: 25 November 1996 16:01

Well, Winter's nearly here, and we need a little controversy <grin>:

So, how about it, is pumping cheating? I'm referring here to true pumping; quickly sheeting in, and then out to encourage "hysteresis" in the air flow on the leeward side of the sail. I'm not talking about "flapping," or "ooching," wherein brute strength is used to move the boat forward in dead calm. Pumping, in most sailing rulebooks, is illegal, beyond 2-3 attempts, typically used to get a boat up on plane. My question is this: Are these anti-pumping rules themselves illegal/unfair? Is pumping simply one more method of using "The natural action of wind over water?"

I'm not looking for a long-winded rehash of rules definitions (although, feel free if you like), but a basic question, "Pump, or don't pump?" I know that normally, pumping is only considered advantageous on small boats, and in limited conditions, but read on. I think maybe not.

Many of you won't be surprised that I'm taking the position that all four of the questions in my subject line are, or should be answered, "yes."

First, Legality. If no more crew muscular effort is put into the system than 1) that used for ordinary control forces, or 2) less muscle in than wind energy out, then it *is* (or should be) legal, and unlimited. For readers certain that hauling in that mainsheet is just the same as "flapping," I refer to Tony Marchaj's new "Sail Performance," International Marine, 1996, ISBN 0-07-040250-7, Chapter 3-6, "Pumping," and also 3-8, "The Katzmayr Effect." (I just got the book, and like any good thriller, it reads best from the back forward ;-)

Marchaj insists (and I agree) that a system could easily be made wherein the mainsheet tensions are trivial (for example via a fully-balanced rig; like a balanced rudder). Muscular force would then be tiny, and the "brute strength" argument goes out the window. Also, such a rig would make pumping possible on far larger boats, and in a wider range of conditions than currently thought possible. (How to build a "fully balanced" rig? A solid wingsail, pivoted about 20% of its chord aft of the leading edge is one approach. Another is the Aerorig, where the mainboom extends forward of the mast, to take the foot of the jibstay. The whole rig then rotates; the jib foot moving to windward as the main clew moves to leeward. Other methods certainly are possible; these are used just for visualization of the forces)

Given no need to "horse" the mainsheet, should pumping be legal? Does it use, "the natural action of wind over water?" Of course it does, it's as natural as birds flying. Of course, rules still need to outlaw flapping and ooching, and sometimes pumping is hard to distinguish from them, but the offending Skipper knows, and rules should be able to be written to follow.

Fair? Well, this is a harder one. I suppose that if pumping is legal for one boat, and is used to advantage to win races, then everybody else will have to follow. I see nothing wrong with this. The same thing happened with asym chutes, fully battened mains, heck, even spinnakers themselves were originally "rule-cheaters." If a class seeks to stifle innovation, and freeze its development at some predetermined point (which far too many do, IMO, even though they insist they don't), then pumping wouldn't be "fair," I suppose. For all others, it's at least as fair as allowing high-tech sailcloth, composite masts, or planing hull shapes.

Advantageous? As it stands today, pumping is generally used off the wind, to get the hull on a plane. It isn't generally used in very light winds (too little total power available to justify the effort), very strong winds (the amplitude and frequency of the pump would have to be great, and the rig is usually powered up anyway). It isn't used when hard on the wind (though I believe it could be), as either the rig is powered up or it's thought that bringing the boom in further than it is will just increase heeling moment. It's not used to leeward, when huge 'chutes are flown instead.

In general, all of these limits revolve around the (current) need for significant force on the mainsheet to initiate pumping. Given a system which minimizes control forces (as above), I believe pumping could be used in many more conditions.

At higher windspeeds, when the boat is not overpowered, if the rig were pumped, it would have to be done at high frequency. This is perfectly feasible, of control forces are small. As, theoretically, pumping increases the rig's coef. of lift, and not the coef of drag, pumping could add thrust beyond that of a larger rig, without increasing heeling.

Just the opposite happens in ghosting conditions. Currently, low speed pumping and "flapping" are hard to distinguish from each other. Pumping a balanced rig, at very low windspeed, would be both advantageous, and easy to distinguish from cheating (pumping would require small amounts of muscle, flapping much larger). With little force needed, long tacks could be "pumped" at all times.

When hard on the wind, with boom amidships, little pumping "amplitude" is available. Bringing the boom to weather, thus bringing the sail force resultant abaft the beam *will* only increase heeling force, but, again according to Marchaj; specifically pumping, and generally sailing in fluctuating winds will, on average, increase the sail's drive force (coefficient of lift) without significantly increasing it's drag (coefficient of drag) (This is called the Katzmayr Effect). Thus the rig averages a higher L/D, and this could/should be capitalized on for greater speed through the water to windward and/or higher courses sailed.

Last, off the wind: This is here where pumping could bring a huge difference. Imagine the whole rig; mainsail and chute, being pumped as a unit. All the increase in rig drive (lift and drag) can be utilized. I see difficulties with pumping a conventional chute, or even an asym, but the results could/should be worth it. Marchaj suggests that a sail's coefficient of lift might be increased as much as 50% during a pump. I wonder what happens if the boat is intentionally driven in a zig-zag course, in order to pump the rig? The total effect, including the resultant pumping of the keel and rudder, gets complicated. It would have to be understood quite well in order to do it to advantage.

Automated? Here's where we go out onto the boat-design limb a bit. If you've bought into the argument so far (and you *should* have ;-), the next step is to contrive to pump the rig *all the time.* A carefully controlled, continuous pump would allow the wind to reattach itself to the sail's leeward side just in time to be pumped away again. Low speed off the wind requires slow, high amplitude pumps, while high speed upwind legs need short, fast strokes. (BTW, this isn't my idea, either. It's from Marchaj)

How to automate this? Given a balanced rig (let's take the solid wing analogy; just like a balanced rudder), one could move the pivot point even further back, and end up with a rig which was dynamically *un-stable.* It would want to slam hard over whenever it tacked. This is ordinarily considered very dangerous, but let's say we could control it (variably raking the entire wing, relative to the pivot point, would be an easy method of doing this. Rake it forward, and the wing over-controls; raking it back under-controls it). Now, it's only necessary to devise a system which quantifies how much amplitude and at what frequency the pumping needs to happen, and we're all set--the rig pumps itself. Microprocessors and airflow sensing devices come handily to mind, but I rather think that a system which responded to rig forces and accelerations themselves could be designed. High force and fast movement of the boom = high windspeed, and vice-versa.

Or perhaps (and more likely) the helmsman would retain control of both amplitude and frequency, in addition to average angle of attack. I have no idea how best to accomplish this, but several routes come to mind (both mechanical and electronic). I kind of like an old AYRS trick of mounting a small tail boom, with smaller wing, behind the main solid wing. This technically makes a ketch or yawl rig, but in reality, makes the rig behave like an airplane, turned on it's side. The tail plane, on its boom, serves to control the angle of attack of the main wing, without the dangerous "over-control" attitude above. Relatively small control inputs to the tailplane result in large fluctuations in the main wing's angle of attack. (BTW, AYRS certainly is not the first, or most recent group to sugext this. There's a version in L.F. Herreshof's "Common Sense of Yacht Design" (I believe first published in 1946). Also, Christopher Hook designed such systems in the 1960's)

So, what do you think?

 

From: FISHWICK Simon Date: 25 November 1996 17:31

Not having read Marcaj's latest, I'm not actually sure what the physical mechanism behind pumping (and the Katzmeyr effect) is. Can someone please enlighten me?

"Flapping" I take to be the acceleration of airflow by physically pushing it towards the stern of the boat - the sort of thing that can be done quite happily when there's no wind.

What's "pumping"? Are we talking about the exploitation of transient effects whereby a sail can be sheeted beyond it's stall angle and, for the few moments it takes for the leeside airflow to separate, so create ultra-high lift coefficients?

Second question: How does this relate to what Bethwaite calls (if I remember rightly) "fanning"?

3rd question: How does one distinguish between pumping, and the rapid continuous rotation of the rig to produce magnus effect lift (cf Flettner rotors)? (other than by the amount of rotation)

From: Dave Culp Date: 25 November 1996 21:30

>What's "pumping"? Are we talking about the exploitation of transient
>effects whereby a sail can be sheeted beyond it's stall angle and, for the
>few moments it takes for the leeside airflow to separate, so create
>ultra-high lift coefficients?

Precisely.

>Second question: How does this relate to what Bethwaite calls (if I
>remember rightly) "fanning"?

Don't know, I'll have to check Bethwaite. Thought "fanning" = "flapping."

>3rd question: How does one distinguish between pumping, and the rapid
>continuous rotation of the rig to produce magnus effect lift (cf Flettner
>rotors)? (other than by the amount of rotation)

Interesting question. Both schemes contrive to disrupt and control the airflow at greater and greater distances from the wing surface, in order to increase the lift coefficient.

I would say that, in the case of a Flettner-type rotor, continuous and significant power must be put into the system, in order for it to work. In the case of pumping, much smaller input (conceivably taken from the airflow itself) is needed to initiate and continue the effect (if the rig is designed for minimal input--see the original post).

You posit an interesting question, though: *If* Flettner rotors were allowed in racing (perhaps through some scheme for self-rotating the cylinder), how would the "sail area" be measured? Surely not the projected area of the cylinder, or even half its circumference. The lift coef. multiplier is just too great.

So; how to "measure" a "pumped-rig" boat's "sail area?" I don't know...

From: PETER_RAYMOND Date: 26 November 1996 02:27

Something like a Flettner type rotor will self rotate. I guess I am easily amused because I thought this was pretty interesting when I first stumbled across it. If you take a piece of balsa wood 1/16" X 1" X 36" (I may not be remembering width and thickness correctly) and, holding it in the center, give it a little flip while launching it, you can get it to auto rotate as it flies. The balsa bends enough that there is a little dihedral and the resulting "flying wing" flies fairly straight. I remember a glide ratio of about (3 to 5):1 at best, which is really bad for such a high aspect ratio, but the lift coefficient must have been pretty high because it flew fairly slowly. On the other hand the lift coefficient may not have been that great since I recall that adding even a little weight allowed the thing to do great dropped rock imitations.

I thought about trying this out as a sail on an 18" balsa catamaran I built a long time ago, but I would have needed to sail it in enough wind that bearing friction didn't stop the rotor and I doubted I could keep it upright with an aspect ratio of 36 and enough sail area to move the boat.

I haven't looked at any variations at all. A different shape for the leading/trailing edges, an X profile, a slightly S shaped profile, a drum with paddles, end plates, or some other variation might improve performance. I think that the wind power people have vertical shaft designs that could probably be modified to produce Flettner effects as well. I've also wondered if you could make a kite that flew this way. Has anyone ever heard of one? Don't self powered pseudo-Flettners sound like an interesting winter project?

>Surely not the projected area of the cylinder, or even half its
>circumference. The lift coef. multiplier is just too great.

Lift is great, but the drag is high. I don't think a self powered Flettner would work very well upwind, if at all. IMHO It's main application would be for slow moving boats where the apparent wind doesn't move too far forward. Kind of a small sail area, high tech square rigger.

>So; how to "measure" a "pumped-rig" boat's "sail area?" I don't know...

I wouldn't penalize it until someone gets it working. Don't impede progress until it's too late.

[This message spun off a correspondence on Flettner Rotors - see below. Ed]

From: Theo Schmidt Date: 26 November 1996 16:06

I think in the context of traditional boat racing it is sensible to discourage pumping. However in the AYRS I think we should be thinking more openly. I would like to see races with open class boats where everything is allowed with the exception that any natural energy is not previously stored (wind and solar power), i.e. any batteries discharged before the race. Human power would be allowed even though this is pre-stored. Such races would be very interesting in areas where the natural conditions vary and allow different boats to excel in different conditions.

From: Dave Culp Date: 02 December 1996 08:27

Please forgive me. This is a long post, and responds to half a dozen people's postings. I've been out of town for a few days, and thought this would make more sense than lots of short answers:

> Scott
> Did you see the biceps on the Olympic gold medal-winning woman? Her upper
> arms were about the size of my thighs!

>David:
> [...etc...long winded justification deleted...]
> If it is not trimming, it must be pumping. Why not just tie a paddle to your
> belt and do the macarena on the stern? It's dancing not sculling, no?

>Dan:
>The rules normally say something like 2-3 times per wave. The reason
>for this is to prevent a rowing contest. I would expect if the boat
>is stopped allowing 2-3 fans of the sails as well as the tiller beyond
>that if you can't get sailing again it turns back into a rowing contest

>Jean-Paul:
>NO PUMPING

These are examples, IMO, of the "knee-jerk" reactions. You guys are saying, "I don't like what I see out on the water, so instead of thinking about it, I'll just vote 'NO'." Please read on...

>Malcolm Osborne:
> Pumping has been legalised for sailboards. Also for this mode of
> sailcraft, as you say, pumping is not advantageous in strong wind, but it
> has the detrimental efect of races being sailed in no or very little
> wind, where most of the motive power is supplied by the sailor. The race
> order is then largely determined by physical strength and stamina. If it
> is legalised for other sailcraft, wouldn't the same sorry state of
> affairs apply?

1) It is not *physically* possible to man-handle larger rigs; clearly it is only *just* possible to do it with windsurfers. Thus there's no need to outlaw it in bigger boats, physical limits prohibit it.

2) "Pumping" (as I've defined it before, and as differentiated from "fanning," or "flapping") does *not* require large amounts of physical energy. You *can* think smart, not hard, and use pumping (true pumping) to enhance your rig's power. This can be done by ordinary humans, even women and children, and adds no muscle power to the boat's forward drive.

>Fishmeal:
> Yes, transient effects are almost certainly under-exploited in many modern
> rig designs. But for conventional soft rigs, I think you'll find that
> most sailors want to play a version of the game that does not involve
> very much pumping. There's no objective argument to support this - it's
> just a question of the particluar skill set that we want to test and
> develop when we go racing.

To the point and on-topic, thank you ;-). However, we *all* sail in transient effects, *most* of the time. Short term (<10 secs duration) wind changes are *far* greater than we normally concede. Marchaj shows direction changes of +/- 20 degrees, and velocity changes of +/- 10%, even for these very short-term time periods. Our orderly minds simply prefer to "average" out real physical events into nice tidy "steady-state" sailing. IMO, we actually *damp* the transient events, to our detriment.

Better to investigate transient events, and seek to use them to our advantage. As to sailors' "wanting to play" a version of sailing pretty much "like our fathers did," this is simple inertia. Multihull sailors overcame this many years ago, and even monohull sailors are learning to use full battens, no?

>Art Engel:
> If you want to argue that a slow "pump" has purely aerodynamic effects
> and because of its speed cannot convey any energy from the pumper's
> arm to the sail then you are theoretically correct, it should probably
> not be illegal. Moreover, you might have a pretty good argument that
> it is not covered by the anti-pumping rule.

I do not seek to circumvent current rules, or try for "theoretical" correctness. I'm looking to re-write the rules altogether. IMO, it's anti-development rules, of which anti-pumping is just one example, which are suffocating sailboat racing.

> However, the practical problem is how in the heck do we tell a
>"non-energy transferring" pump from a normal "pull the boat faster
> through the water" pump. We really can't so both end up getting
> covered (though as a practical matter continually adjusting the sail
> is consistent with your "slow, non-energy transferring pump" and might
> not get you a protest).

This is a valid point, and deserves some study, not simply "throw the baby out with the bathwater." Further, perhaps I was imprecise; I wasn't only proponing "slow" pumping. In many instances, pumping needs to be rapid and vigorous, yet still needn't involve "muscling" the boat forward. Again, an example: If you build a rig which is "balanced," and has no "feel" in the mainsheet (mainsail area ahead of the mast, through one of several mechanisms), it would be perfectly possible to "pump" the rig, even in very large sizes, without large amounts of muscular energy going into the system. One way of "defining" pumping (versus "flapping") would be self-evident, at least with larger rigs. The crew's muscles simply wouldn't be up to the task of "flapping" a large sail; yet could easily "pump" even a very large, though balanced rig.

Of course, other methods of differentiating between toe two modes is necessary, for smaller classes of boats. Marchaj says, and I agree, that the perpetrator knows, *absolutely* which he is engaged in, as it takes a radically diffeent "feel" to the rig to do one than the other. I expect rules can easily be written to differentiate, as well.

>> Art Engel (again):
> > It is interesting to note that windsurfing has eliminated the
> > anti-pump rule and apparently the result is that the "sailors"
> > with best weight-training programs consistently finish in the
> > top of the fleet.

This is curious, to me. What you say is true, although generally, windsurfers aren't actually "flapping" their way forward (though sometimes they do...). Generally, they are legitimately "pumping" the rig (by my definition). However, their muscle power isn't going towards forward momentum, but rather to overcoming the imbalance in their rigs. They're constantly overcoming mainsheet tension (or its equivalent), then giving this energy away by sheeting back out.

What surprises me is this: If pumping is so obviously advantageous with boards, then why hasn't an enterprising designer worked out a balanced rig, in order to make pumping easier? The potential is certainly there, and such a rig could still be "flapped," should that motion be wanted, too. Maybe I'm posting to the wrong list of sailors, no? ;-)

>Jim Champ:
>I guess my first thought is that continuous effort on the sails all
>the way round is somewhat inelegant.

And yet, *especially* in light air, we all seek to play the mainsheet (and jibsheet, and 'chute) pretty much all the time. To take this thinking to it's logical conclusion, we should just cleat the sails down at the mark and live with the consequences...

>...mega snip...
> >>Automated pumping?
>Intriguing - and if it could be made to work mechanically and without
>electronics I suspect it could even be legal. Bearing in mind the
>complexities though I don't know if it could be done without
>electronics...

Let me ask you this: Are wind vane self-steering gears legal in racing? I know nobody uses them, except offshore, but are they legal? (I don't know) How about electronic autopilots? I'm not asking of they are a good idea (they're not), only if they are legal. The answer would be the same, wouldn't it, for automated pumping gear, either electronic *or* mechanical?

>One wonders if anyone in the leading edge of the
>aerospace industry is looking into these effects, could it provide
>more lift and a lower stall speed? I don't know what the passengers
>would think if the wings oif their 797 started flapping on final
>approach though:-)

Go rent "Top Gun," and pay particular attention to the shots, taken backwards along the jets' fuselages, back towards the aircraft carrier. There are lots of these throughout the movie. Pay attention to the jets' control surfaces (flaperons, rudders, whatever they're calling them now). These puppies are *literally* flapping. They're responding, via computer control, to just such transient events as we're talking about.

>Fishmeal again:
>This prompted a continuing debate with my fligh
>instructor about whether pumping the flaps up and down by hand could
>marginally improve performance in the final stages of a too-low dead-stick
>approach. I still maintain that it's *possible* for some judicious use of
>hard flap pumping to make the difference between "just clearing the fence"
>and "buying the farm."

It *absolutely* would. Go back and punch him in the mouth ;-). I'll happily go up with you and help get some data points, to prove your hypothesis. We'll bring Tony Marchaj along, too.

Return to Table of Contents

From: Robert Cavenaugh Date: 02 December 1996 15:59

Quite aside from the question of legality, it seems there may be a flaw in the proposal to 'pump' a substantially balanced sail or rudder. While either works reasonable well while sailing, it seems to me that pumping would upset the balance of that system. Any area before the mast (or before the rudderpost) would pivot in the opposite direction from the aft portion, and would therefore pump in reverse, spoiling the desired effect, and perhaps even stalling the boat's forward motion.

An unbalanced sail involves higher sheet loads, but if you pump it you are delivering all the force of the sail in the desired direction. Or, at least, so it seems on this dark morning....

From: FISHWICK Simon Date: 02 December 1996 16:34

Assuming

1) pumping to be "the generation of short-term (transient) increases in sail lift force (above steady-state values) through changes in sail angle of attack" (NB pumping needs an pre-existing wind across the sail)

2) "flapping" (Dave's term) to be "the generation of increases in sail thrust by using the sail to impart extra energy to the airflow across it" (can be done when there is no wind, or when the apparent wind is dead-ahead)

3) - most important - that pumping as defined above is possible, and produces useful increases in drive

Then

One might expect the production of a pumping effect to be VERY dependent on the local air velocities across the sail, and on the relative velocity of airflow and sail surface.

In which case, how effective pumping is may depend on

1) the point about which the sail is rotated

2) the rate of change of angle of attack

3) finding an optimum (2) for a given (1)

[the idea being that if the sail luff moves downwind too fast the airflow in its lee is disrupted almost immediately (ought to be visible as luff collapse with a soft sail), whereas if it is moved more slowly the airflow is maintained, and produces lift beyond the normal stall values (for a while . . .) - IF that's what is happening]

Also

4) there is a severe risk that the whole problem is not analytical without resorting to a CFD simulation!

PS With a balanced rig, flapping may not be possible - the leach producing forward drive being offset by the luff driving backwards - think of a symmetrical sail pivoted along its line of symmetry. . .)

From: Kristy KML64 Date: 02 December 1996 17:39

No one has really addressed pumping the spinnaker on boats like the J22 and J24. We have found that the big boy trimming the chute can pump it to jump up on a plane whereas the smaller person can not react fast enough with the use of a winch, it has to be a strong person who can do it with brute strength. Is this cheating? Just curious.

From: Dave Culp Date: 02 December 1996 22:49

>Robert Cavenagh:
>Any area before the mast (or
>before the rudderpost) would pivot in the opposite direction from the aft
>portion, and would therefore pump in reverse, spoiling the desired effect,
>and perhaps even stalling the boat's forward motion.

You're presuming that such pumping would be fast enough that the fwd portion of the foil (air or hydro) would move faster than its local windspeed. This is unlikely, under any conditions. Even a languid 5 kt wind involves an airflow of, what, 10 m/s. You'd need to haul the boom end in at 3-4 times that speed, just to get the forward portion (typically 20% of chord) moving fast enough even to see zero-knot local wind velocity. This high speed of rotation probably isn't possible; it certainly is far faster than is wanted, in order to instigate the hysteresis effect we're looking for.

>An unbalanced sail involves higher sheet loads, but if you pump it you
>are delivering all the force of the sail in the desired direction.

You're still looking at pumping as "propelling" the boat forward. It's only a rotation of the sail, resetting its angle of attack, that's wanted. You don't "pull" the boat forward (at least, not legally ;-)

Simon Fishwick:

>Then
>One might expect the production of a pumping effect to be VERY dependent on
>the local air velocities across the sail, and on the relative velocity of
>airflow and sail surface.

Certainly true

>[the idea being that if the sail luff moves downwind too fast the airflow in
>its lee is disrupted almost immediately (ought to be visible as luff
>collapse with a soft sail), whereas if it is moved more slowly the airflow
>is maintained, and produces lift beyond the normal stall values (for a while
>. . .)

See above response to Robert. The rotational speed just isn't that great; after all, we *want* the lee side flow to remain attached. We don't want to just yank the sail surface away from it.

>Also
>4) there is a severe risk that the whole problem is not analytical without
>resorting to a CFD simulation!

Windsurfers seem to have managed to find the "groove" by simple iteration. The "proper" angle, frequency, rotational velocity, etc. is the one which causes the vessel to move ahead of its competition... ;-)

From: Dave Culp Date: 02 December 1996 23:43

Robert Campbell, from private e-mail:

>You are welcome to quote
>any of it if you want to respond on the list. (i.e. if you think
>there any valid points that further the discussion).
>Is it possible that many people misconstrued your use of the term
>"pumping"? I'm not sure if I've got the jist of it correctly either. Or
>do we (as a group) misunderstand what pumping is? I think, because you
>have said 'as differentiated from "fanning," or "flapping"' that I do
>understand it, but I'll continue anyway. (I don't have Bethwaite's book
>to refer to, so I don't know how he defined the various terms.)

Yes, it's particularly hard to hold a proper argument without a clear definition of what we're arguing about ;-)

I'm not referring to Bethwaite's book, but rather to Marchaj's latest, "Sail Performance." Sounds like you have "Sailing Theory and Practice." Still a good book. "Sail Performance" isn't the only place to read about pumping, though none other comes immediately to mind...

I probably made a bad assumption; not that more posters had read Marchaj's latest, but that more had had previous debate or experience regarding pumping, beyond the "I don't like cheating" sort.

>so that when you rapidly crank in on the main sheet one could imagine
>it "pushing" the air back to propel one forward. Is this an
>oversimplified view of what pumping does when one is trying to get up on
>a plane or to catch a wave?

This is "flapping," not "pumping."

>Is it really just that one is altering the
>angle of attack momentarily (as someone else mentioned) to give an
>increase in drive and not really giving a "push" in the sense of trying
>to displace a volume of air rearward?

Precisely. The airflow on the lee side of the sail doesn't detach, for a few moments, but continues around the lee side, in normal attached flow. You thus have, for a very few moments, the mainsail sheeted at a very high angle of attack, with commensurate high lift coefficient (plus no great increase in drag). It's as if you could extend the L/D curve, on it's normal slope, far past the stall point. (Rather, it's not "as if." You actually *are* extending the L/D curve.) After a bit, the airflow breaks away, drag goes through the roof, and you have to sheet back out to re-establish normal flow. *Nothing* unusual happens on the windward side of the sail. No air is "pushed" to leeward, etc., during true pumping.

To take a "macro" rather than "micro" look at things, the whole sail system has encouraged a far larger volume of air to disrupt its journey, and bend to the boat's rear. This momentum transfer is what powers all sailcraft. Bend a small amount of air = small amount of power. Bend a lot of air (smoothly, now, without creating energy robbing eddies or vortexes) and you can steal away a great deal of energy.

>I think we tend to think of pumping as my former description
>and not as a rapid trimming.

Those who don't do it, or don't do it effectively, certainly do. I'm trying to encourage those who've *never* tried it to think about it.

>It certainly isn't a trimming because of
>a rapid momentary shift in wind as the sailor isn't aware of that change.
>In fact would an instrument be aware of it, given the effects of wave
>and other changes in boat direction on the apparent wind?

The sailor is most certainly aware of which he is doing. Pumping gives rather great amounts of added power, compared to energy in. You get the definite sense that things are multiplying. "Flapping," or whatever you want to call it, is definitely just "rowing the boat." The successful pumping skipper knows the difference. If you want to try it out, go out in a small dinghy in light wind. You can flap in no wind at all, pumping requires airflow, which you are (gingerly) manipulating...

>It seems to me though, that there would be inertia in the rig. Also, the
>rapid change in attitude of the rig (assuming something like the Balestron
>rig) would change its apparent wind wouldn't it? So say you detect
>sudden header of 10 degrees, so you crank in the sheet, but the rotation
>of the rig would mean that the apparent wind at the forward end of the rig
>would have shifted even further forward, no?

I hope I've answered this in an earlier post to the list. Ask it again if I haven't.

>I have trouble imagining any sort of soft rig responding quickly enough
>for this rapid pumping idea. Perhaps it is appropriate for a hard wing
>rig though. Just something for you to think about. I can see Dave's next
>rig now: a hard wing kite sail. Might hurt when it is time to bring it
>down though! Even a hard rig would seem to have too much inertia, unless
>the pumping could be achieved just with flaps.

See the same post. Keep visualizing the sail as moving *slower* than the local airflow. "Backwinding," even with a hard wing, would be counter-productive.

From: PETER_RAYMOND Date: 02 December 1996 23:53

I doubt that the rules on pumping will change much, and I don't like to stand in the way of progress, but:

>Robert Cavenagh said:
>>An unbalanced sail involves higher sheet loads, but if you pump it you
>>are delivering all the force of the sail in the desired direction.

Dave Culp replied:
>You're still looking at pumping as "propelling" the boat forward. It's
>only a rotation of the sail, resetting its angle of attack, that's wanted.
>You don't "pull" the boat forward (at least, not legally ;-)

I understand that a stationary boat in still air will still require effort to rotate the sail, even around the center of area. But, I think the force required would be quite a bit less than is typically required to "pump" a sail by any definition I have heard so far. To me it's easy to assume that zero effort pumping isn't very likely, so looking at this from a conservation of energy point of view I conclude that all pumping includes an element of flapping. I think that CFD would probably show the same thing, dynamically changing the angle of attack adds energy to the air stream and it comes from the arms of the pumper. I would like to see a comparison between the efficiency of pumping vs the same energy expended pulling on a rope attached to the boat just ahead. Does anyone have a feel for how this comparison would turn out based on personal experience?

>Simon Fishwick:
>>Also
>>4) there is a severe risk that the whole problem is not analytical without
>>resorting to a CFD simulation!

Dave Culp:
>Windsurfers seem to have managed to find the "groove" by simple iteration.
>The "proper" angle, frequency, rotational velocity, etc. is the one which
>causes the vessel to move ahead of its competition... ;-)

Based on limited observation, and some inept personal experience, "proper" pumping for windsurfers has a fair amount of motion of the aft hand and some motion of the forward hand as well. I would say that the center of rotation is at, or ahead of the sail's leading edge. My most vivid memory was a case where pumping pulled the board up on to a plane when the wind was light enough that the boards would never have gotten there without pumping. Final apparent wind was probably about 15Knots and once on a full plane pumping was no longer needed. The pumps were fairly vigorous and had as much pull as rotation. I would guess that that real world pumping includes significant flapping and I don't see any way to separate the two, even considering more balanced rigs.

From: Kristy KML64

No one has really addressed pumping the spinnaker on boats like the J22 and J24. We have found that the big boy trimming the chute can pump it to jump up on a plane whereas the smaller person can not react fast enough with the use of a winch, it has to be a strong person who can do it with brute strength. Is this cheating? Just curious.

From: Dave Culp Date: 02 December 1996 22:53

I would tend to say it's not cheating, I'd expect others to say it is ;-)

Perhaps if you'd experiment with easing the guy at the same time as hauling in the sheet, you'd find that less energy is needed. You're then rotating the sail around it's center of mass--sail plus sail-full of wind--rather than accelerating all that mass of enclosed air.

Most skippers would likely consider all this a waste of talent, and disrupting to his tactics, at least until the boat shot forward and left the competition in the wake...

From: Dan Frenette Date: 03 December 1996 00:03

Another low energy thing you can do is run a line through a block midships to a block floating on the spinsheet (well on a small spin like the J22 maybe a loop is enough). Pulling this hard will open and close the top of the sail so you get a good puff on the top of the main where you want it to drive the bow down. Also when the boat kicks off on a wave jam this line in and it will tend to pull the spin flat and this will have the same quick trim but about 1/2 the hard and you get back to loss trim much faster.

As far as skipper goes if they know what you're up to they should ignore it and sail the boat and let the trimmers deal with course changes etc.

From Kristy: KML64 Date: 03 December 1996 01:01

Yes, it will certainly shoot you forward . . . and I guess that's why we put the big ones on the sheet and the little one (me) does the tactics.

Our method is to leave a little curl so there is something to "pop" rather than "pump." It is amazing how it works.

I'm going to try easing the guy next time as you suggested and see if I can get the same effect.

From: Dave Culp Date: 03 December 1996 07:41

>Peter Raymond:
>I understand that a stationary boat in still air will still require effort
>to rotate the sail, even around the center of area. But, I think the force
>required would be quite a bit less than is typically required to "pump" a
>sail by any definition I have heard so far. To me it's easy to assume that
>zero effort pumping isn't very likely,

I could easily build a sail system which was "over-balanced." I'd do this by pivoting the sail *behind* it's center of pressure. It would now want to slam hard over, without any input from the helmsman, beyond bringing it back to the eye of the wind (which I'd cleverly manage to do also, perhaps by moving the pivot point forward again, the important point being that I'd use the power of the wind alone, not the helmsman, nor the trimmer.) Such a system could easily pump away, forever, without any human energy input at all.

Marchaj's thinking, his "proof" that the helmsman's energy input isn't important to the success of pumping, goes thusly: Take an ordinary mainsail, and pump it. Measure the energy input, and the increase in the sail's drive. Now, move the pivot point aft, closer and closer to the sail's center of pressure, and run the experiment again. You'll find exactly the same hysteresis effect, and the same increase in drive, but with *less and less* power input from the helmsman, as the pivot point moves back. Thus, it's demonstrated that the energy derived from pumping is independent of the energy input by the sailor. (with apologies to Tony Marchaj for the paraphrasing).

>The pumps (with a Windsurfer) were fairly vigorous and had
>as much pull as rotation. I would guess that that real world pumping
>includes significant flapping and I don't see any way to separate the two
>even considering more balanced rigs.

It is considerably more difficult to "flap" a rig which doesn't either a) have a universal joint at the deck or b) is attached to a hull which can be rocked by the crew; either motion is necessary to "flap" the sail. Simply "fanning" it back and forth won't move the boat (not much, at any rate). Another poster accurately characterized the motion as more "scooping" the air, not "fanning" it.

Pumping, on the other hand, has none of this "scoop." The sail needs to rotate only about a single axis. Thus, pumping is possible on large boats (if the rig is balanced), flapping is not.

From: Wayne Laurents Date: 03 December 1996 08:01

Pumping on a sailboard (Windsurfer, Mistral, Tornado, whatever), I would describe as sculling through the air. Feather the sail, tip it forward, flare it(as in "sheet it out"), then pull the rig back strenuously. I have tried not pumping in regattas, and been severely stomped, but when I pumped, I could (and several times, did) win. Works in glassy calm.

From: FISHWICK Simon Date: 03 December 1996 10:07

On my Mirror dinghy, I had an endless spinnaker sheet/guy which allowed the sail position to be controlled independently of the sail shape. (Think of it a running through two block on the quarters in a "W" shape, with the middle point of the W passing though a floating block. Adjusting the position of the floater adjusts the sail draft. Moving the whole sheet through the blocks adjusts sail trim to wind. NB the tension on the guy is always more than that on the sheet, so the sail tends to "bear away"and stall if let go)

With that kind of arrangement it ought to be easy to check if momentarily rotating the rig (paying out guy) followed by a slower return produced a significant pulsed increase in drive - and it ought to be proof against over-pumping protests! ("All I did was let go, y'r Honour")

I don't know if it will work on anything bigger than a dinghy though (the forces may be too great for easy manual handling)

Previous Page | Table of Contents | Next Page


The AYRS Newsletter is published quarterly by the Amateur Yacht Research Society Ltd, BCM AYRS, London WC1N 3XX, UK. The AYRS as a body is not however responsible for opinions expressed by individual authors.

This publication is copyright under the Berne Convention and the Universal Copyright Convention. All rights reserved. Apart from any copying under the UK Copyright, Designs and Patents Act 1988, Part I, Section 38, whereby a single copy of an article may be supplied under certain conditions for the purposes of research or private study, by a library of a class prescribed by the Copyright (Librarians and Archivists) Regulations 1989:SI1989/12, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means without the prior permission of the copyright owners.

Multiple copying of the content of this publication without permission is always illegal.

Authorisation from AYRS to photocopy items for personal use, or for the personal use of specific clients, is granted ONLY to libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, CCC, 21 Congress St, Salem, MA 01970, USA. This authorisation does not extend to other kinds of copying, such as for resale, or to other unregistered organisations. Permission to copy articles for these and other purposes should be sought on each occasion from AYRS. Permission for non-commercial purposes will not normally be refused.