Recognising seamanlike rigs
There are lots of designs for sailing rigs and they cannot all be seamanlike. I need to interpret that word “seamanlike”. There are rigs that have evolved to a point where the users of the rigs are satisfied that they are as good for their intended purpose as they are ever likely to be. The dhow used on the Nile, the Norfolk wherry and the Thames sailing barge, with its Dutch equivalent, are typical cases. They look to be primitive when compared with a modern racing yacht. But they are all working boats that have always had to earn to continue in existence. They are cheap to build and are operated by a minimum of crew and they work in special situations. Any unbiased assessment will recognise that these boats and their rig are seamanlike. They work.
By comparison there are other designs of rig that seem to have come about just to be different like putting the mainsail in front of the jib or for some special purpose such as sail training when a hybrid rig is mounted on a hull. Some designs seem to spring from optimism.
I argued for the flow over the windward face of a sail being orderly and free from eddies and swirls and that this is the best that can be done. I came across the picture in figure 87 of a “super” yacht. It is a radical design. There must have been very good reasons to set the sails in front the mast and disturb the flow on the windward side. I cannot think what they might have been.
I discussed the mode of operation of a single soft sail and a Bermuda rig. In figure 88 we have two small boats both of which are racing and presumably going as well as they can. They are beating in the River Fowey but probably not quite as close to the wind as they might be. We can see where the single sail of the Topper is set to the wind and I have to suppose that this is the best position, though perhaps not the best shape, for this isolated sail. On the Troy the mainsail is set with its main boom a few degrees off the centre line. If it had been the only sail it would be in the same position as the sail on the Topper so it is not set to drive the boat but it is in the best place to interact with the jib to let the jib drive the boat. This is to be expected from the physics. Even though the design of the Troy has been unchanged for 100 years it looks to be quite modern and I do not think that anyone would regard it as unseamanlike. (Troy's have no inbuilt buoyancy and they do sink from time to time but they are recovered.)
We can see a compromise that has been taken in the rig on the Troy. The boom must have a kicking strap and must swing across the cockpit freely. This requires space below the boom for the kicking strap and the sailor and the boom is fitted some way up the mast. But the jib is low down and seemingly not in the best position to take advantage of the flow round the main. However there is the vortex coming off the foot of the jib to be accommodated so the unavoidable gap may be useful.
However one of the small dinghy classes, the Feva has a very exaggerated vertical misalignment of the two sails. One wonders whether there is any interaction at all between the main and the jib. Indeed I am told that it uses a gennaker to get round the fact that the jib does not pull its weight.
If the Bermuda rig has been developed to improve the upwind performance of a yacht we should be able to see it at work. In figure 89 the yacht is beating and we can see the angles of the two sails. They are consistent with the jib working in the diverted flow in front of the main. Note that the jib is set low to close the foot of the sail with the foredeck and the also that there is space under the main to accommodate the kicking strap and provide working space for the crew. The rotating flow coming off the main goes though that gap.
If the Bermuda rig involves setting the jib at an angle to the main for beating what about rigs with two or three jibs? The main and the inner jib will, between them, divert the flow
more than the main by itself. So we might expect to find the forward jib at an even greater angle than the inner jib and that is what we have in figure 90 where the yacht is some way off the wind. The interactions still give an advantage over the single sail at this point of sailing.
Figure 91 is of a catamaran that is beating. It has two mainsails and at first sight it seems that they are not interacting yet closer inspection shows that the forward sail has more twist in it than the aft sail. The two jibs are set just as might be expected and all four sails are acting together as a rig. When I first looked at this rig I thought that it was too far forward relative to the hull but recognised that I did not know the position of the keels. However there are two wakes that appear to be coming from the keels and the position of these wakes suggest that the boat would not be balanced. Look at the wake off the two rudders; they are being used to prevent the boat turning to leeward where a balanced boat would tend to turn to windward. The designer of this boat may have good reasons for this set up but the drag from those rudders must have been a significant factor in the design compromises.
Figure 92 shows the yacht on the cover of this book. It takes this rig to new complexity with three mains and three jibs all fore-and-aft. The interaction between the three mains appears to be in a progressively larger twist in the sails which is much easier to achieve with sails that are gaff rigged. The inner jib looks to be not in the best position but the other two jibs are where they might be expected to be.
If you accept my argument for the interaction of sails then you can also see when a rig is “wrong” and the inner jib should be in a position that is consistent with the rest of the rig. However there may be some good reason for this.
In figure 92 the yacht has furled square topsails. It also has furled top-sails on each mast. I do not understand the mechanics of square sails used in this way and I can see no way to find out what they do and how they do it.
I have looked at lots of pictures of boats under way and they all seem to fit in with this idea of making the sails interact to get the jibs into the best position to drive the boat. I have also seen some that seem only to be comprehensible if the basis of the design is to fill in every available space in the standing rigging regardless of any interaction. I have not saved pictures of such rigs.
I have often thought about the Bermuda rig and wondered just how it could be made to work for the best. This text provides an opportunity to see what might be done to disentangle all the conflicting factors that make it difficult.
I need some sort of overview. In essence my argument is that in order to improve the upwind performance over that for a single sail the sail area of the rig can be shared between two sails, the main and the jib, and the jib can be made to work in the diverted flow ahead of the main and be better placed to produce a force that acts forward to drive the yacht.
In figure 94 I have brought forward a diagram from section 3 on the Bermuda rig. Let me recapitulate. The diagram shows a single sail with its boom lying along the centre line of the hull. It also shows the flow pattern round this single sail when it makes the angle to the apparent wind that would be appropriate if it were to be the main sail of a Bermuda rig. However there is nothing in the Bermuda rig that requires this particular arrangement, the boom could be set to leeward or, indeed, to windward if its interaction with the jib when beating is improved.
For the given position the diversion of the flow ahead of the sail is evident and there is a circle that shows the likely place for a jib. It is clear that the useful area is smaller than the area affected by the main sail by a diameter ratio of about 2 to 1. It follows that we must expect the jib to be smaller than the main.
Now look at the diagram in figure 95 brought forward to re-examine. I drew the deck with the main set on it as it would be for beating. Then I added a jib in an arbitrary position in order to show the typical arrangement of the forces on the two sails. Now I want to reopen the location of the jib.
For the given position the diversion of the flow ahead of the sail is evident in figure 94 and there is a circle that shows the likely place for a jib. It is clear that the useful area is smaller than the area affected by the main sail by a diameter ratio of about 2 to 1. It follows that we must expect the jib to be smaller than the main.
Figure 95 has been drawn with the jib mounted on a boom at its foot with a swivel joint between the boom and the deck. The swivel has been placed at about the quarter point on the boom so that the sail will automatically tack when the yacht changes tack. The position of the swivel on the deck is on the centre line and as near to the mast as permits the boom to clear the mast. This is a common arrangement but fitting the jib on the forestay at its luff is even more common as my pictures all show. Generally the sail is located to suit the mechanics of the yacht and not to give the best performance when the yacht is beating.
I think that some progress can be made towards finding the best position for the jib if alternative positions for the jib ate found without reference to the practical implications. In figure I have given the position of the jib for the luff attached to the forestay in red. I have also given a new position for the jib in green. When I drew this diagram for the first time I let the booms be parallel for all three positions but the prospect of having a useful interaction between the sails looked to be compromised and so I let the booms have angles that look to be at least likely to give sensible flow pattern.
In chapter 9 of the book on RC yachts in this website there are pictures of flow patterns in a Hele-Shaw apparatus showing the effect of changing the angle between the jib sail and the main sail and they show that the flow bends upwards even more than it does for the main to pass under the jib and that there is a best gap between the leech of the jib and the luff of the main.
It is clear to me that there is a case for the jib having a best position when its luff is well to windward as shown in green and some new angle to the wind that is in a new position that is better for the direction of the net force on the jib to drive the yacht. However the implication is that the swivel on the boom is now nearer to the rear quarter point and the sail will not tack automatically.
It is the insert in figure 96 that is interesting. There I have drawn the net force on the jib in each of the three positions and also the component of each force parallel to the centre line. The green component is 20% greater than the red component. This is an important improvement because it is the jib that is driving the yacht and not the main.
I do not think that the physics is incorrect and one might now think of designing some mechanical arrangement to have the swivel back on the forward quarter point and the swivel moving across the deck during a tack. When I look at the rearrangements that are made to set large spinnakers it seems to me that some device to set the jib in what could well be a better position is at least worth some thought. But it is never going to be a simple device and it is sensible to look elsewhere.
Knowing that turning the jib through an angle can be so useful one must go back and look for other arrangements. I have redrawn the flow pattern round a Bermuda rig to clarify the flow lines. It is figure 9. It shows that two sails work together to produce a new and different flow pattern for that for a single sail.
Compare it with the pattern in figure 6 especially in the region bounded by my circle.
Figure 97 is for a rig with one jib having a wide chord and it is difficult to fit it into the flow pattern. However if one thought of using two jibs that together had much the same area as the single jib each would have a much smaller chord. Then the inner jib could be much closer to the main sail and work in a flow that has been turned as far as it can be and so set the force on the sail in a much better angle. This inner jib with the main would turn the flow ahead of it further again and the outer jib will work in a flow that has been turned more that the main can turn it. The two jibs are a much better proposition than one jib and they give considerable freedom of design. We can see the result in figures 3, 4 and 5. Examination of figure 8 shows that the flow ahead of and in the vicinity of the mast is robust and in no way dependent on a knowledge of the flow over the lee side of the sail. An inner jib having a small chord could be placed in the sharply curving flow and provide a more curved flow for a larger outer jib. This is practical because jibs of more or less equal size are in widespread use. The physics hints at a refinement being possible.
If the observer is actually aboard a moving yacht this graph is distorted by the motion. This could be modelled but it is more than is justified here. So let us suppose that the yacht is beating on the port tack. The angles of attack of the sails will be close to the minimum relative to the apparent wind. If now the wind swings clockwise it will reduce these angles and if no action is taken the sails will collapse. The skipper has a choice, to bear away or to tack. If he bears away the speed-made-good will drop but if he tacks the speed made good increases. But when the switch occurs the crew must be ready to tack again to gain another advantage.
How nice it would be to know what was in the wind. I understand that a helmsman may be able to tell when a wind shift is coming from the behaviour of the boat and its response to the helm.
Off shore racing often takes place in wind that has flowed over the land and the topography of the land can either be known to the skipper or simply observed from the yacht. There must be links between the behaviour of the wind that flows over the yacht and the land over which the wind flowed to get to the yacht. One must form a mental that relates the two and then refine it by feeding in all sorts of other information that comes with experience. Having an understanding of the physics of airflow is a great help in gathering data in a reliable way and being on the lookout for information at all times is very rewarding.
The most important step is to get a good idea of how the air flows round obstructions and of the scale of any disturbance in the wind. By that I mean that an eddy might be a mile across or a half a mile and the size at a given site will depend on the size of the obstruction and might depend on wind speed. It must be handy to look at the coast and the speed and direction of the wind and have a good idea of the character of the wind in which you are sailing. That information can be taken on board by the skipper of a yacht and used to advantage.
Just looking at the flow of water is always instructive and it can be sought out. On Par beach in Cornwall a swiftly flowing stream runs out to sea on the Eastern side of the beach and creates a new bed in the sand after each tide. The creation of that bed is a quite rapid process and, in a very short time, the water flows round ever-changing shapes in the sand. Given a dog to walk trying to interpret the flow in very rewarding. It can be transferred to the flow of air over hills and through valleys. It is all an ongoing process for those with the will to see. It is a way of learning to read the wind in the absence of innate natural ability.
Using my models of sail action and wind shifts
For those who are not born with the innate ability to “read” he wind an understanding of the physics of the flow of fluids makes learning about the wind much easier. The explanation of veering and backing above is just part of the mental model.
I thought that I might be able to detect the passing of a vortex using a kite. I bought a kite and flew it on a beach at about thirty feet. The first vortex that passed lifted the kite and swung it to the left then back again to the right and finally to drop back to the middle again at the original height. It was a gust. By chance the second vortex let the kite drop, swing to left and right and back to the middle and then climb back to its steady position. It was a weakening of the wind. I have to say that I was pleased to observe this behaviour although the lifting of the kite in the gust took me by surprise. I had thought that the kite would let me see the veering and backing and overlooked the possibility of seeing the gusting.
This paper has addressed wind shifts but the free stream of air contains all sorts of disturbances that are not organised into vortices. How does the sail cope with these? The soft sail can only operate when it is stalled and the flow has separated from the convex side. The convergent flow on the concave side quickly suppresses disturbances and the flow over the convex side cannot be affected because it is already eddying vigorously. So the soft sail can cope with randomly disturbed flow. It is extremely well suited to its job. However the use of a gaff permits much better control of the twist of a sail as is very evident in the case of the Norfolk wherry and that twist makes the sail even better at coping with disturbances in the wind.
 I am sure that there are readers who regard theory as suspect at the very least and think that empirical methods are much to be preferred. It is my experience that the judicious use of physics, if you can find it, together with empiricism is a quicker and surer process to reach a sound position to start design. I am writing a textbook to support this contention and it is on this web site.
 A cross between a spinnaker and a Genoa.