The absence of test data for sails

I was reminded today that I am intending to add the attachment given here to my website. It occurred to me that some members of this forum might have something to add. So, at the risk of being shouted down, I offer it for your attention.

 

Good thinking in your attachment.

 

I have a lot of aerodynamics experience in aircraft design, and have also done a lot of sailing; I have similar observations as you noted about the lack of understanding in the sailing community of how the sails actually work. There are good scientifically done test at various universities around the world, so there are some that do know the issues affecting real sails. However you have a number of social factors to overcome to get alternative sail designs accepted by the sailing public. And you have I think a flaw in your underlying assumptions, that there is a strong market demand for better sail designs.

One problem is that since sail boats have been around since the dawn of recorded history, clearly you can get even poorly designed sails to drive a boat before the wind. So many designs evolved by chance and luck, with very poor terminology and understanding of the forces at play. By the time aerodynamic theory was developed enough to allow the invention of aircraft, you also had efficient engines that made commercial use of sails obsolete. That means that sailing is now largely a recreation, driven by tradition, marketing and consumer desires. Most sailors want a sail boat to look traditional, hence little significant change in recreational sailboat designs for the last 100 years or more. There is little to no commercial value to developing better sails designs, so it is only for recreation (consumer driven) or racing (rules driven) that new sails are now developed, each having severe limitations in what is desired by consumers or allowed by rules.

I have found it almost impossible to have an intelligent conversation with experienced sailors about sail design, because of the many misunderstandings, misnomers, and ignorance of how sails actually drives a boat. There is really little benefit to attempt to get sailors educated about what would work best, since there is little commercial benefit to it. Sailing is just a hobby, sometimes a very expensive one, but there is very little financial benefit to better sails design, so there is not a lot of expensive research that goes into it as compared with say cars, aircraft, medicine, etc. Almost all of the advancement is sails has been as the results of technologies gained from other industries, and almost no original developments, but borrowed technologies and evolution in designs as consumer desires have slowly changed.

So other than a hobby to satisfy your own curiosity, what benefit is there in developing better sails? Sails that few if anyone would want, and sails that most racing organizations would quickly ban (as "unfair"). Good luck with that.

I am writing this as an engineer that likes playing around with sail and hull designs to make the boat sail better and faster. Perhaps some day I will invent something that everyone with a boat will want, but I would not count on it. I am content with just messing about with boats for my own interest and enjoyment.

How would you be able to use the knowledge of how to make a better sail design?

 

Ivor,

Having found your website, I will need time to study it and perhaps see how your thoughts vary from what I have gleaned from other books and papers plus experience over the years. One principle question is; have you devised and built a sailing rig that performs significantly better than dozens of existing rigs or, do you find a jewel among developed rigs that is significantly better than commonly used ones?

Well, that is actually two questions but at least they are connected.

 

The absence of test data

Please see attachment. I do wish that I could paste into this window.

Ivor Bittle

 

Ivor,

First, you can paste your word processor document into the reply-to-thread window. You write your responses in Word (unformatted) and then pick "select all" from edit menu (or right mouse click), then pick "copy", and switch screens to the response window after you get on-line, and then right mouse click "paste" in the window. Your text will appear in the reply window.

Your observations are correct, the jib acts similar to a leading edge slat on an aircraft wing. It is known that most of the thrust or lift comes from the jib/Genoa, and it is known that as much as half or more of the main is not generating any lift because of poor shape, mast interfearance and other flow problems. Hence the development of larger and better jib/Genoas. This also means that much of the main sail only adds weight and drag, so there is much improvement to be made.

It is also found in testing that the same principals of aircraft theory apply to sails. End plates, elliptical plan form shapes, high aspect ratio, streamlining the mast, all improve the sail's efficiency. they have even tested shaped battens and/or various sail tensioning schemes to get more optimum sail camber and shapes. Oddly most of these known improvements have never made it into recreational sailing, and many have been used in racing sail boats that do not have as restrictive rules (but these classes are pretty rare).

Both wing theory and flight and wind tunnel testing confirms that a high aspect ratio single surface produces the best L/D ratio, yet there has been no development of large single surface sails (a modern cat rig). All of the fastest sail boats use sloop rigs. I think that most of the interest has been on optimizing max lift from the rig, rather than max L/D (hence the focus on the jib). It seems to me that refining max L/D would get you less heeling moment, which means you can have less weight on the keel or in catamarans and much less total displacement. Better L/D would also mean higher speed for a given wind speed, vs. going for max lift, at any cost in terms of drag.

If you were to develop a more efficient rig the mast shape must be part of the "foil" profile, and it must be allowed to pivot, and the sail must be fully battened with variable stiffness battens (along their length) to optimize the sail camber. Everything should be done to reduce sail and rigging drag (eliminate wires and spreaders, go to streamlined struts, use smooth textured sail cloth, etc.). High aspect ratio, an "end plate" at the top of the mast (like an airplane "winglet") and an end plate boom (or have the boom brush across the top of a flat top cabin roof to act as an end plate-tests have shown this is effective). If you were to use a jib (I am not convinced yet these are necessary with a well designed mast/sail combo) the jib should also be high aspect ratio, and it can not have the jib form a point at the top. And they will also need battens to maintain the best camber shape. This means they can not be roller furled.

By the time you are done messing with so many aspects of the sail, it will not look very "traditional" and will likely have poor consumer acceptance, and there are few racing classes that allow such extreme changes. In the rare few unlimited classes they are already using rigid "wing" sails with triple slotted flaps, LE devices, full ribs and D-section leading edges. Very revolutionary, but not very practical for daily sailing.

I am presently working on a sailing rig concept that might allow similar performance without the complexity, and it has the ability to reef or remove most of the "wing" area when the fabric part of the sail is lowered in the ordinary way. It has the ability to accurately adjust both the camber and the twist all along the length of the sail with a relatively simple control scheme. It should also be much lighter than a similar sized conventional rig, it is boomless and has no cables or spreaders. I hope to build an 18' catamaran hull in next year or so to try it out.

 

Petros,
Very interesting. How could I have acquired that information in so direct a form except through a forum? I will think about it and reply properly by pasting. My thanks.

Ivor Bittle

 

The conditions that a sailboat faces in reality are seldom as simple and straightforward as those we use to generate our theories. For this reason, there is always great room for interpretation of observed results. The best L/D ratio will not always give the best performance. Often, the greatest driving force will be best even if L/D is low. Other factors like stability of the boat, wind pressure and hull drag can affect which rig is best. A modern jet wing like a 727 operates with L/D ratios from very low to very high depending on whether it is taking off, landing or in high altitude flight. We can only approach this versatility with wing sails and flaps but have not yet developed effective leading edge slots for boats.

The history of C Class catamarans probably best illustrate the evolution of sails where only the total sail area is restricted. They have evolved from sloop rigs to an articulated single wing sail. Even here, the necessity to sail in all directions relative to the true wind greatly affects the shape of the optimum sail so that "one" best shape will not be dominate. Unmeasured jib overlap area plus the ability to carry huge spinnakers off wind keep the sloop rig dominant in racing circles. Even the newer breed of sport boats stay with the sloop rig even though the uni-rig is best in terms of L/D. A much lower CE of the sloop may also get the boat around the course faster than a high aspect ratio, high L/D uni-rig.

It is just too simple to disparage the "racing crowd" for retarding new developments although it is often true. The fact that development classes have not made great or spectacular deviations from one design class sailing rigs should offer some evidence that the standard sloop rig is pretty good all-round sail configuration.

My favorite rig is the unstayed cat-ketch. It is reasonably close to the others in performance and enough superior in handling keep it at the top of my list.

Gotta go do something useful.

 

One word: Marchaj.

 

http://na.northsails.com/racing_sails/gradient_series.html


World-class sails, Quantum Sail Design Group (Annapolis)
http://www.marylandresearch.umd.edu/issues/spring2003/business.html

MIPS-developed technology is used in nearly all of Quantum’s sail products.

Quantum created the first U.S.-based wind tunnel testing facilities for downwind sails with the University of Maryland's Glenn L. Martin Wind Tunnel through MIPS.


http://syr.stanford.edu/SAILFLOW.HTM

http://cat.inist.fr/?aModele=afficheN&cpsidt=2745537

http://www.wumtia.soton.ac.uk/yachttesting.html

Just a guess here, Ivor, but I'm thinking that between all these research groups, there's more than a mountain of stable data from which to draw. Now, getting access to it may be problematic, but it is out there.

 

My thanks to recent contributors. I will make time to read and think carefully and then reply as necessary. To Amolitor I can only say that if you are happy with Marchaj carry on. When I read Marchaj my reaction was to think that I can do much better than that and I has been working on it ever since. It have given me immense pleasure that you will pass up.

Ivor Bittle

 

Other books to read

Don't forget Frank Bethwaites articles and books like High Performance Sailing. Frank has a wealth of experimentally backed theories for his and his son's rigs. One of Franks predecessors was Manfred Curry. He has books published from the 30s. Frank has a new book in production. It has wind tunnel test data and lots of theoretical grist to chew on.

I second the bit on the real world getting in the way of ideals. Little, small boats gets so knocked around it is almost impossible to known what the flow conditions of the air in the rig is. C Class cats sail in smooth water and are big so their rigs seem to be able to get close to theoretical ideals. For me I think that you will be a very long time modelling the turbulent behaviour of a small dinghy rig in chop.

cheers

Phil

 

Best of luck to you, Ivor! Marchaj did a lot of excellent work, but science always allows the next generation to advance another step, right?

 

I consider myself fortunate that the first book I read on theory and tactics was Curry. It was published in 1926 and gave a foundation to judge other works and ideas against. Amazing insights Curry had. I also obtained Marchaj early on and devoured both books. I've read Bethwaite and think he is an original and creative thinker also.

Still, there seems to be room for more work on sails in the future. I doubt that sail design has kept pace with hull design in the area of high speed performance. Not so long ago, 10-15 knots was considered very high speed for a sailing vessel. Now boats make 24 hour runs at speeds of 25 knots or more. Multihulls are at the front of these speeds but it is no longer proper form to call monohulls dull "lead mines" as they are making incredible speed also.

 

I cannot take on board all the comments to this thread. Suffice to say that they have added to my working knowledge and I am grateful.

Ivor Bittle

Chris Ostlind points me towards the website of Stanford Yacht Research. Go there and read the first section on Basic Theory (of Sailing). I have no idea what you might think about it but as an engineer I find it falls a long way short of what I expect in this context. How can I have any confidence in anything that might follow if they can sleep peacefully after posting this as their opening statement?
If you go on with this website you will find the outcome of CFD applied to sails. I think that I got closer with a pint-sized Hele-Shaw apparatus that I made at home and used in the garden taking pictures on cold days using a first generation digital camera. If you want to see the outcome look at Chapter 9 of my book in Section 4 of my website. http://www.ivorbittle.co.uk/. Pictures from the Hele-Shaw equipment have to be interpreted. In order to do that you need to understand what is going on. When you have done so at least the data has been obtained from a real piece of equipment with real water flowing in it and the outcome may not be relevant but it is real and can only be challenged on the grounds of poor experimental technique.

This goes to show just what a mountain there is to climb if we are ever to have a rational explanation of sail action.
The contributions to my thread have been very useful. So useful in fact that I think that I shall either have to reconstruct the whole document or ditch it.

It seems to me that the basic problem hinges round the character of the flow over a sail. I can conceive of no way that an un-battened sail can ever have attached flow on its leeward side. This, in fact, is the reason why the sail is so successful. Only a very few aeroplanes are designed to fly in both the un-stalled and the stalled condition and those that are have to be handled carefully as they go though the stall. Changes in modes of flow are very troublesome and to be avoided if at possible. The sail is always stalled and has no sudden changes in its mode of flow and this single fact makes its use over the ages reasonably predictable and safe.

If you deny that the sail is stalled and proceed then any theory that may be developed is pretty pointless. However I would defend both my right and your right to be hopelessly wrong. No reader should ever drop his guard and just accept what he reads.

If you accept that the sail is stalled then the role of the jib of the Bermudian rig becomes obvious and it is easily seen to be the equivalent of the slat used on a wing. It cannot be part of the main like a slat because the main is stalled as is the jib for that matter. The jib is there to extend to windward the working range of the main operating alone and take the yacht closer to the wind. And that raises the question of whether the best design for the main should be the one that gives the greatest diversion of the upstream flow (to assist the jib) for the least drag. It does not matter much if this makes it less “efficient” at other points of sailing, it is pointing that is important.

Let me digress for a moment. Container ships require their metacentric heights to be controlled between fairly close limits and the magnitude of the metacentric height is really very small when compared with the main dimensions of the ship. The position of the metacentre is known for a ship when loaded to its marks. What is not known is its centre of gravity. It seems that the containers are loaded in some controlled fashion so that a computer can track the final position of the centre of gravity. In order to do this the weight of each container must be known and also the position of its centre of gravity.
Measuring the weight is not a problem but the centre of gravity certainly is. Opting out of this problem is not possible. The ship must sail. A strategy has been adopted in which the position of the centre of gravity is arbitrarily ascribed to each container according to its content and weight. Initially the arbitrary positions erred on the safe side, and, after a period of experience the positions were adjusted to give a better value of the centre of gravity. We are part way through the whole process. There is a clear strategy here and there are other strategies in engineering to solve difficult problems. Almost no one identifies these strategies yet, if we did, we would have a set of standard strategies and be much better able to cope.

Now there is a need for a strategy to cope with this thorny problem of gathering experimental data and I began to outline a possible strategy. There I did not include what engineers call good practice. Petros made a comprehensive list of items that seem to me to be the essence of good practice. With his permission I shall incorporate it in my final text.

My attention was drawn to pictures of yachts being tested in wind tunnels and these convince me that we desperately need some general scientific principles to govern the use of combinations of sails into rigs. Organisations are proud to show photographs of models under test when there seems to no way that the sails can work together. Do they have no way of selecting likely designs from the possible designs or are they under contract to test whatever the customer dreams up?

But there is a feature in all this that bothers me. People enjoy “messing about with boats”. If they all took my clinical view of sailing they would not do this and life would be not so rich as it is now. Perhaps, as Petros says, sailing is recreational and best left to “granny logic” and enthusiasm. No one is going to try to get to the moon using sails

In his first note Tom 28571 makes several pertinent points and then says that he was off to do something useful. His second is a powerful observation of the current scene but made me wonder where he placed the astonishing performance of windjammers.

Amolitor puts a rhetorical question to which the answer is no. Everyone wants to just continue from where we are. We need a few to take stock of where we are and suggest the best way forward. New Scientist just recently noted that when computer modellers came up with something that did not fit the facts they tended to refine their models not reconsider the basis of them. They behave like socialists who, when something does not work, apply more socialism. That is the way of life. Marchaj achieved his objectives in writing his books. Now we need some one to take the whole affair forward again.