Section 6  Other aspects of the Thames sailing barge

 

Contents

 

Introduction to appendices to do with full-sized barges

  Appendix 1 The mizzen sail

 

  Appendix 2 Screw steering on a Thames sailing barge

 

  Appendix 3 The top sail “tensioner”

 

  Appendix 4 Capsizing incident

 

Modelling appendices

  Appendix 5 Tools for making models

 

  Appendix 6 The case for rudder only

 

  Appendix 7 Setting up a Spektum computer controlled transmitter for controlling the sails of a bowsprit barge

 

 Appendix 8  Pictures of barges under way

 

 

                                   Scroll to search

 

 

 

Introduction to appendices to do with full-sized barges

The Thames sailing barge has proved to be much more complex than I first thought. At various times I have pondered on some aspect of the barge only to find that I did not understand it. There are things like the mizzen sail and the steering gear that took a long time to sort out and there is still the system that tensions the top sail that I still do not understand with total confidence. At various times I have written my thoughts on these things and I have brought these together to make this section.

 

I shall call them Appendices and give then a number and title.

 

Appendix 1 The mizzen sail

I thought that I had sorted out the little mizzen sail used on spritsail and bowsprit barges. Then Peter Mortimer asked me to say where he should fit his fin and lead. This simple question opened up the whole problem of the little mizzen again.

 

The question is still, “what were they used for?” It seemed to me that there are almost too many different bits of information to fit together. Let me start at the beginning with figure 1 due to March. He claims that this is the early (1820) form of the little mizzen that he calls a jigger. The sail is mounted on a mast that is on the rudder-post and is stayed forwards to the tiller. It must have been used on a small barge. The sail has a sprit and a boom. A sail with a boom normally needs a kicking strap but in this case the boom is pulled down by a rope and tackle attached to the blade of the rudder. What is so special about this jigger?

 

It seemed to me that its primary function would be to help in a tack. In 1820 barges were propelled by sail, by the tide, and by men with oars and winches. They worked up all the rivers. Sooner or later they would have to tack up a river such as the Thames at Gravesend. Tacking is always a tricky manoeuvre and especially so in a tideway. There are two features of the jigger that seem to me to be important other than the fact that it is the only sail that I know that can be set to windward. The first is that the sail appears to have been rigged to be flat by having everything tight. (I checked this with Bill Sutherland and he says that they were very tight like a board and so does John Bannester.) The significance of this is that the sail cannot collapse and just flutter as any sail having camber can collapse. This means that it can work usefully at small angles to the wind. The second is that the sail will turn with the rudder. Now suppose that a barge is to tack. Bill Johnson says that it will be made to bear off the wind a little to pick up speed and then the rudder put over to swing the head through the wind. The success of the tack is now dependent on the speed of the barge and the effectiveness of the rudder. Bill Sutherland says that a laden barge is very slow to turn as I should have expected. It could all go wrong as model barge sailors know only too well. But, if the jigger sail goes over to windward with the rudder, it will help to “weathercock” the stern round and, because the sail is flat, do so for longer than a cambered sail could. The jigger can help in a tack.

 

But jiggers evolved into little mizzens. What was so desirable that they had to change? I am coming to think that two or three factors came together to produce a major change in the sail handling of a barge. Let me list them.

 

1 A barge is a big boat for a crew of two or even one man and a boy. It is bigger and much heavier than the largest mono-hull used in single-handed ocean racing. It needed to be balanced in some way to hold a course by itself. (A sailing boat is balanced when the sails comprising the rig are, for some course, set so that they drive the boat efficiently and need no rudder offset to hold a course.)

 

2 The steering mechanism, whether chain or screw, was so simple that, in Bill Sutherlands words, “a couple of turns of the wheel were needed to take up the slack”. How could a barge hold its course with a steering mechanism like this? (Any slack in the steering of a car shows up alarmingly.) I now think that the screw-operated steering gear is fundamentally unsound and it could only work with lots of slack as I have explained in appendix 2.

 

3 Buried in the design of a sailing barge is a compromise. The length of the hull has to be apportioned to the principal parts, ie the holds, the rig and the lee-boards. As the rig and the lee-boards need a space all across the deck that is a bit more than one third of the way from stem to stern, there has to be two holds, one small one forward of the mast and one large one aft of the mast. Some of the length is needed for accommodation, fresh water tanks, chain lockers etc but these do not dominate the arrangement and are accommodated in the run aft and the shaping at the bow. I do not know how a barge was loaded and unloaded but it seems likely that most of it was man handled. It seems likely that small holds meant climbing ladders whereas large holds might involve ramps. No doubt all sorts of cargo-handling methods evolved and best practice came to be established. Then some general conclusion would have been reached about the best arrangement of the holds and the rig within the length and the space of the hull. It looks very much as though the decision involved a main hold that was as long as possible. As the sprit had to span the length of the hold to use it as a crane-jib this meant having a mainsail with a long foot. The outcome, rather to my surprise, was the standardising of the size of the deck hatches.

 

Figure 2 is the sail plan of SB Kathleen as she was originally rigged. She is typical and obviously it is a very practical arrangement. However, if the mizzen were to be removed, as in figure 3, the rig would seem to be too far forward. If the figures are run out for the sail plan less the mizzen, the rig is not balanced but, if the mizzen is included, the sail plan balances in the middle of the submerged area of the lee-board when it is in the position shown on the plan. This corresponds to almost exactly 40% back from the stem. (When Kathleen was re-rigged as a staysail barge the rig was re-balanced in the same position.) The conclusion must be that by the time that Kathleen was built the jigger had become an important element of the sail plan and was in fact a mizzen sail.

 

So the mizzen looks to be a busy sail. It needs to have sufficient area to balance the rig. It needs to help in a tack. If possible it should help with the steering. How did it manage all these things?

 

Figure 4 shows two barges racing. They are beating close to the wind. I was surprised to see that both rudders are off to leeward. The enlarged view of the sterns in figure 5 show that they really are to leeward and producing quite a wake as a result. I looked at the other pictures in March and all the barges were the same.

 

There must have been a significant advantage to these two boats that were racing to justify accepting this unwanted drag. I think that the little mizzen and the rudder acted in opposition to give self-steering. Let me explain. No sailing craft can beat or reach without having its hull turned to windward relative to its course to give it leeway. Clearly the rudder in the picture must be trying to turn the hull to leeward when the hull must turn to windward. It follows that the mizzen must be overcoming the action of the rudder and turning the hull to windward.

 

Now a barge will have to sail in winds that gust and veer. I am fairly confident that gusting and veering are linked. I think that the wind veers increasingly to one side as the wind speed increases and then switches quickly to the opposite side as the wind speed is at its maximum and then the speed falls back as the direction returns to normal. During this process it is likely that only the mizzen is directly affected. As the wind gusts and veers, the force on the stiff little sail increases and pushes the head up into wind which effectively increases the angle of the rudder which then resists the mizzen to limit the change in course. As the veering and the gust dies away the reverse process takes place to bring the barge more or less back on to course. It is a self-steering mechanism that can be set up by setting the rudder to leeward.

 

But this system solves another problem because it gets rid of the effect of slack in the steering by biasing the rudder to leeward where is pushes in only one direction on the screws. Of course the slack will appear again during a change of tack but, once the change is made, the slack is taken up again.

 

It would be nice if it all stopped here but refinements were made that involved moving the mast off the rudder and on to the after deck and increasing the size of the mizzen sail. It looks very much as though the little mizzen sail was harnessed to drive the barge to make up for the drag from the rudder as well as its other functions. The sheeting became controllable by the helmsman and this little sail became a maid-of-all-work under the control of one man. The Thames sailing barge seems to be full of surprises.

 

It still had one inherent sang. When a barge goes into irons it stops and, with no flow over it, the rudder can have no effect. The mizzen is still in the wind and still active. The normal way to get out of irons is to sheet in and wait. Ultimately the barge will fall off the wind, the angle of the wind on the sails will increase, and at some point the sails fill and the rig starts to drive the boat and that brings to rudder back into action. However the mizzen is waiting and will try to turn the barge to windward before the rudder develops its full effect. The barge cannot get out of irons by this method unless the mizzen is eased off.

 

Some barges were fitted with an automatic sheeting system for this mizzen. It seems that chains of equal length are attached to the aft end of the rudder and made off to the transom so that the chains make about 45° to the transom when the rudder is on centre. These chains are long enough to permit full rudder travel and consequently they sag. If the rudder is, say, turned fully to windward the leeward chain will straighten and lose its sag and the windward chain will sag still more. If a rope passing over fairleads on the transom is connected to the centre points of the chain it will be hauled to windward as the rudder is moved. If the chains are heavy enough this movement can be used to move the boom of the mizzen and adjusted with a block. It is automatic for tacking and the block can be used to adjust the sail for tacking and possibly for reaching and to let the sail go for beating.

 

Finally, with the conversion of barges to have engines, barges motored up rivers and tacking became unnecessary. The mizzen became part of the sailing rig and now comes in a variety of sizes to suit the rig. John Bannester says that the change took place after 1975.

                                                                              

I am indebted to Bill Sutherland and to Bill Johnson both of whom have extensive experience with barge sailing for letting me question them to see whether this explanation of the mizzen fitted with their recollections. What they said convinced me that I am on the right track now. Indeed it fits together so nicely that it leads me to marvel at the people who designed, built and developed these boats. I would be pleased to hear from anyone who has anything to add to this brief article.

 

Appendix 2 Screw steering on a Thames sailing barge.

Some long time ago I was pondering the way in which the steering appeared to be used on a spritsail barge and asked Bill Sutherland, who has experience of handling barges, how much free play there was in the steering wheel. He said that it was as much as two turns. My engineering ears pricked up and I wondered how this could possibly come about. I suspected that, if this mechanism was constructed properly, it could not have so much free play and that it would jam if the rudder were to be turned too far. I checked and the steering gear is well made.

 

The diagram shows the mechanism as drawn in The Handbook of Sailing Barges. There is a substantial shaft with robust opposite hand, two-start, threads. Nuts run on these threads and, through arms, move the ends of the rudder bar. It looks all very simple and I have seen versions of this in drawings of other types of ship.

 

However, right from the start I have had doubts about the viability of this mechanism It is asymmetrical and it looks wrong to me and I thought that it would jam if turned too far. Yet it was and is in widespread use. All the time I was supposing that the shaft ran in bearings and was located end to end by collars or the like. Then, I was looking at the diagram above and noticed that the rudder end of the shaft is supported in a bearing in the rudder arm and evidently can float, at least at this end. I looked for a restraint elsewhere and could not find one. I concluded that the shaft must float axially.

 

This put a whole new slant on this mechanism and led me to analyse it by computing (See end of text.). Sure enough the shaft must float if the mechanism is not to jam. But the price to be paid is that it is a mechanism with no triangles[1] that is very vulnerable to any wear or free play and will magnify any slack. I thought that I knew where Bill Sutherland’s two turns of slack came from.

 

When the calculations are run for typical dimensions for the steering gear it shows that, with no slack anywhere, the shaft must move about 1 inch when the rudder turns through 45° before it jams. This 1² was going to be visible so there was only one way to find out what was correct and I went to Maldon to look. Immediately I saw what I needed on Hydrogen.

 

 

In the photograph we can see the wheel covered up, the shaft in its pedestal bearing, which is clearly not suitable to withstand an axial force, and between them the gadget with a handle. That gadget is a band brake. If the shaft had been constrained so that it could not move axially that brake would be attached to the pedestal but, because the shaft moves, the brake has to move and it is tied to the deck instead. The shaft is rusty between the brake and the pedestal but nowhere else.

 

This just raised another question. What advantage was gained from using this seemingly unsatisfactory mechanism when much more accurate steering could be achieved by doing away with one screw, its nut and the link, and preventing the axial movement?

 

At first I thought that two arms were better than one should one arm break. However this would not do because steering would fail completely if either arm broke. Then I spotted a note that said that rudders were most heavily loaded in a swell. I do not fully understand this statement but it did draw my attention to the fact that, as the rudder is pintle mounted and not underwater, waves of any sort will effectively dip the rudder more deeply and add to the rudder area just when big pressures are being produced by the motion of the water. A rudder that cannot give will be very vulnerable. The single-screw, single-arm and single-horn arrangement has no give but the conventional one does because the shaft moves until the steering gear jams and if the forces suddenly change direction the gear will move a long way before it jams in the other direction. And this may be the whole reason for the retention of this unlikely mechanism. In turn this leads to the fitting of a mizzen and using the rudder as part of a self-steering device and to do jobs that the rudder cannot do with this form of steering.

 

By chance I had an opportunity to speak to Bill Sutherland again and I asked him whether the steering shaft could move. He said yes it did move fore and aft as if everyone should know that and added that the rudder continually slapped when at anchor in a swell and it was often lashed. I find it intriguing that the development of the mizzen sail was necessary to make up for the unwieldy way in which the rudder must be controlled just to stop it being damaged

 

Computation

I have drawn the mechanism in diagrammatic form. The rudder has a steering arm that is in fact two arms joined to a square strap that fits over the rudder post. I have drawn it as one line with each arm having radius . The rudder is turned through angle . There are two links of unequal lengths connecting the rudder arms to two nuts that run on right and left hand threads on the steering shaft. There will be a point  that is fixed on the steering shaft and always mid way between the nuts  and . If the distance from  to the point  does not vary as the steering wheel is turned all is well but the distance changes with the angle  as is evident in the graph. In order to draw this graph I had to give the mechanism some dimensions. I let  =12² and the links have lengths of 30² and 54² and calculated  for  from -45° to +45°.

 

The graph shows that there is an end float of about 1² and that the mechanism is not symmetrical left and right although not by much.

 

 

Appendix 3 The top sail “tensioner”

There is a mechanism that I think acts as a tensioner for the top-sail. I have drawn a diagram of the main-sail and the top-sail as they are rigged. I think that there is a puzzle in this rig. It involves the way in which the sprit is supported. The foot of the sprit is supported by the stanliff, which is a steel wire from the lower doubler, and a fitting called the muzzle that I have described elsewhere. The muzzle swings around the foot of the mast and is linked to a collar on the foot of the sprit that is connected by chain to the stanliff. The sprit is then supported by the topping lift that is attached near to the mid-point of the sprit and runs over a block fixed to the main mast and down to the pulpit where it is controlled by a multi-rope pulley system. In effect the foot of the sprit, when it is supported by the topping lift, is located vertically and horizontally just like the jib of a crane.

 

This is all in order but the main sail is supported between the main mast and the head of the sprit by the head rope of the sail and that head rope is wire and not adjustable for length. Mechanically, either the head rope or the topping lift is redundant. Either rope could be removed and the sprit still be supported.

 

I do not know which rope is the primary support but I do know that the sprit is heavy and its head just swings round the mast and provides a excellent fixing point for the clew of the top-sail. Now I want to describe the top-sail tensioning arrangement.

 

I have shown the essential ropes etc. in red. The sheet for the foot of the top-sail goes from the clew and over a block fixed to the head of the sprit. This sheet terminates in another block as shown. Now comes the interesting party of the mechanism. There is another rope that is tied, as standard, to the stanliff, runs over the pulley on the end of the sheet, then back through a block that is attached to an adjustable fitting on the stanliff and, then, to a cleat on the sprit. The adjustment of this fitting is by inserting a locking wedge. Clearly this arrangement can be altered in several ways but basically, if a force is exerted by the top-sail on its sheet the stanliff will be pulled to aft and behave just like a spring. The more the sail pulls the more the stanliff will be deflected and the greater the resisting force will be. A consequence of this will be that the foot of the sprit will be lifted.

 

This raises two questions, the first is can the stiffness of the spring be adjusted and the second is how can the sprit lift? The second is easily answered in that for a typical rig the lift of the sprit is about one inch when the travel of the sheet is, say, 10². This is not much is a continually flexing rig. It is the tension that is difficult to deal with.

 

The initial tension can be adjusted at the cleat. Bagshaw says that when the best positions for the lower fitting should be marked once the best positions have been found. From this we can infer that the position of the knot is not critical as we might have expected just because it is a knot. This must mean that the primary adjustment is the lower wedged fitting. It also means that there must be some criterion for knowing when the best position has been found. I do not know what that is but my best guess is that the skipper will listen to his barge on the move and eyeball the top-sail and know from experience when all is well. What is important is that somehow movement of the wedged fitting must change the stiffness of the “spring”.

 

There must be a way to analyse the system to find out what it does but the dimensions are too uncertain to permit an accurate analysis. Normally one can get somewhere by making sensible simplifications. I will do just that.

In figure 2 I have drawn the mechanism in a second position in green just to show what happens when the top-sail pulls on its sheet hard enough to make it move. I have shown the new configuration in green. The stanliff has been pulled to aft at two points and the stanliff now three straight lengths. I drew force triangles for the several points on the rope as distinct from the sheet and it was immediately obvious that the cleat is not in the right place. Figure 3 shows where it should be. I must explain the change. The tension in the rope from knot to cleat is the same throughout. In figure 3 give or take a little the force exerted on the stanliff at the lower wedged fitting is twice that at the knot. In figure 2 it is much less. This mechanism is sensitive to the angles made by the rope.

Now I want to look at this idea of the stanliff being used to create a spring. In my view the way to proceed is to look at something that is simple to understand and works in essentially the same way. In figure 4 I have just drawn a single rope that is fixed at the top and attached to a roller in vertical guides at the bottom and is vertical. This is not that much different from a stanliff. I have supposed that a weight of 1,000 lbs is applied at the roller. If a sideways force is now exerted on the rope at some point such as that shown the rope will move sideways and lift the weight. There must, for a rope of fixed length, be some relationship between the applied force, the horizontal movement of is point of action on the rope, and the distance of this point of action from the roller.  That relationship can be found quite easily once it is accepted that the force applied to the rope is equal to the sum of the reactions at the fixing and the roller. Two elementary force triangles are all that is needed. In order to draw the graph I took the length of the rope to be 240² and plotted graphs for several heights for the point of application of the sideways force.

 

Clearly this mechanism can act as a spring and moreover the force is nearly proportional to the sideways deflection. The “spring” gets stronger as the point of application gets nearer to the roller.

 

Given this insight to the real mechanism on the barge it is reasonable to think that as the pull at the knot can be quite high up on the stanliff and the force on the knot is only about a half of that on the lower wedged fitting the position of the knot will not be critical. It is the position of the lower fitting that matters.

 

I think that this explains this mechanism but it is doubtful whether it will have any noticeable effect in a model barge but it functions on my model and it is all part of scale modelling.

 

Appendix 4 Capsizing incident

I was idly reading Carr’s book on barges and stumbled over an account of a capsizing of a barge.

 

 

 

 

 

 

 

 

 

 

 

 

The critical phrase is “-she must be kept sailing-”. To me this raises two questions, first, “is the statement correct?” and second, if it is correct, “how does this come about?”

 

I have to ask how one could find out that a barge that is being driven hard might capsize if it stops when close hauled? I do not suppose that many skippers actually put it to the test. I do not see how one might do a sort of intermediate test when the barge might nearly capsize. It may have been part of sailing folklore although one must be careful not to be dismissive.

 

Let us suppose that the statement is correct. What this account is describing is an occasion when a barge that was almost certainly light for racing was beating close-hauled up to the wind, and failed to complete the tack and lost way possibly to come to a stop. I can do this with a model very easily and reliably at Maldon in a steady wind. I cannot paddle the stern round so I have to wait for the head to fall away to leeward and for the sails to fill and start to drive the barge. Presumably the unfortunate Mr Austin would have done the same but in squally conditions. How did his barge, that seemingly could carry its rig when sailing, capsize when it was stopped or had very little way? I have not noticed an excessive heel at the start of the acceleration phase as the model gains way.

 

On reflection there is a likely explanation. The barge was sailing light and would have had its leeside lee-board down. When sailing that lee-board would have generated a force acting to windward to equal the component of the force on the rig acting to lee-ward. There will also be a righting force being generated by the asymmetrical flow over the bows. Without this force the boat can capsize. Mr Austin let the barge stop, the righting force from the bows died away and then, with his lee-board down, the capsize is inevitable because the more it heels the greater the upsetting moment on the barge as the lee-board goes deeper.

 

Modelling appendices

 

Appendix 5 Tools for making models.

 

I look round my workshop and I have hundreds of tools for every conceivable job that I might want to do. They are housed in a brick outbuilding that is heated to prevent rusting. Tools are so cheap these days that I suppose that everyone has hammers and pliers and screwdrivers galore. Even so there are some tools that are used a very great deal more than others and some that are a bit out of the way but are especially useful to modellers. I list some of them that I have found useful and it seemed to me that my list of such tools might spark others to tell us what tools they find especially useful.

 

 

Tools

1 Since I built my sprit-sail barge I have attempted to show lots of members that men like me can sew bolt-ropes on to sails. Most of them cannot see the stitches! So freshly cleaned glasses are a must especially if you put them on. They are no help in the knick-knack drawer in the dining room. So glasses are the most important tool.

 

2 I think that number 10a blades to go on your Swan-Morton scalpel are my most important hand tool. I buy them in packs of 100 and might use 5 in a days modelling. Working with blunt blades is just plain not satisfying. Remember the old saying that you get cut by blunt tools not sharp ones. This is the rule that makes you decide whether that you would rather be miserable when you are modelling or spend a coin or two and enjoy it. I look for satisfying hours per pound spent and not just hours per pound.

 

3 Modellers do not need their tools to be set up for the woodwork associated with house building. It is better to use a new sharp metal-cutting blade in your band-saw. You need about 24 teeth per inch. Square up the table to the blade. Run it on a lowish speed.

 

4 I would not be without quality piercing saw blades (not fretsaw blades) fitted in a hand fretwork frame. Piercing saws are used by jewellers. They cut metal and wood. If a saw does not cut straight, use pliers to twist it at both ends until it does cut straight. Get a good fret table. Buy a piercing saw frame (like a junior hacksaw frame) to cut brass tubing and small diameter brass rod. Do not put your saw on the bench and then put something heavy on it to break the blade.

 

5 I think that a 12 inch stainless steel rule marked off from end to end in my favourite units 1/10² and 1/20² has proved to be a godsend. It lets me work to 1/100² as standard. I do not like millimetres because they are too small to see easily on a rule and much too small to divide into 10. Lathes and milling machines are still mostly sold in inches so it all works together.

 

6 I bought my sanding block from Hobbies 69 years ago and it is still the best I have seen. It holds the sand paper tightly to a flat wooden surface about 2 3/4² by 5². Mostly I use P60 grade paper for general use and use a brass suede cleaning brush to clean sanding debris from the sandpaper.

 

7 Request a razor saw for father’s day, one with fine teeth. You might find it handy for sawing but you will also find the removable blade very useful as a short flexible straight-edge and a sanding shield.

 

8 62 years ago I bought drawing instruments like draughtsmen use. It is so handy to have quality kit for drawing. I suppose that they might turn up at boot fairs. They don’t make them like that any more.

 

9 There are all sorts of spring clamps but they are none of then as useful as 6 sets of toolmakers cramps. More father’s day presents.

 

10. If money is no object buy some diamond hones for sharpening tools. You will never regret it.

 

Glues

In the good old days glue was made by boiling the bones and hooves of animals. It was melted in a glue pot heated with boiling water and, when set, was strong enough to make very good furniture. I do not suppose that anyone uses it now except for restoration. During the war Aerolite and Cascamite made their appearance and were used extensively because they were waterproof. They came on to the market for general use as did Araldite. They were all two-part glues, were strong, but, unhappily, slow setting. Hull building was painfully slow and involved lots of pinning and waiting.

 

The arrival of cyanoacrylate changed all that. Out went the pins and, using accelerator, strips could be fitted to hulls instantly using the penetrating properties of cyano to get the glue into the assembled joints. Yet old plans still recommend the earlier glues because they were what was available when the plans were drawn.

 

Araldite was superseded by epoxy resin glues

 

11 Get thin and thick cyanoacrylate glues and accelerator. Make sure that you do not inhale the fumes. Work out of doors or run a fan over your job. Get 5-minute, two part epoxy resin. If you can, buy some epoxy thinner to make epoxy varnish from epoxy glue. Buy some cocktail sticks to mix it with. Forget aerolite and araldite. Do not listen to those who say that cyano softens in water, it can be boiled. If your fingers get too attached to your job try acetone to dissolve the glue.  There is also a range of supposedly waterproof glues like Resin W. I used it on my aeroplanes at one time and when it came time to break up wings etc I just left them out in the rain and later lifted out the servo and undercarriage mounts ready for the next aeroplane  Do not dismiss these glues. If you want to use them make sure that all the joints are coated with epoxy resin.

 

Five-minute epoxy and cyanoacrylate makes lamination possible in a sensible time. Roof beams can be laminated from 3 or 4 lengths of 1/32² ply formed over the appropriate radius. Companionway curves can be made from two or three layers of 1/64² ply. I often use balsa laminates. Three layers of 1/16² hard balsa make an excellent ply that can be cut more easily than one piece of 3/16² balsa because the saw does not follow the grain and it is porous to cyano to make strong joints.

 

I think that I can discern three different wind environments in which sailing boats can be expected to operate.

 

Appendix 6 The case for rudder only

 

The first is the open ocean where the wind is generally steady and free from eddies both large and small. One reads of windjammers sailing for days on the same rig setting and presumably the sails are handling the minor fluctuations of the direction and speed of the wind without producing observable effects on the rig. The rig would be set to minimise the demands on the small rudder to hold the course. A well-developed rig would do this and drive the boat as fast as it can go.

 

Around coasts, where a great deal of sailing takes place, the wind can be disturbed by thermal activity (on-shore and off-shore winds). Winds blowing over the land and then out to sea usually are disturbed with relatively small-scale, random eddying and from time to time by a rotor that gives both gusting and veering. The sailing rig will deal with the small-scale eddying but the crew must respond to the gusting and veering if they are racing. The design of the boat would not differ a great deal from that for the open sea. It will have a balanced rudder.

 

Small estuaries and the larger rivers produce the worst conditions. Inevitably the wind approaches over the land but has no distance to travel before reaching the sailing boat. Disturbances have too little time to die away or to become organised into vortices. The rig cannot absorb these eddies yet there in no time for the crew to respond. A significant uncertainty is inherent in this wind. I do not know of any special changes in design to alleviate this situation and I think that development of technique might be the only hope. The smaller boats will have unbalanced rudder and tillers.

 

There is a group of sailors who depend on technique and they are the ones who sail on rivers such as those crossing the Broads. They must sail in winds blowing over the land, relatively narrow waterways and must continually change course to follow the turns of the river. One of their aids is a wide unbalanced rudder operated by a tiller. It can be used to paddle the boat to change course. They also use what seems like excessive twist in their sails. This reduces the drive as such but extends the range of angles of the wind for which the sail produces this reduced drive. Add the fact that these boats have very little parasite drag on their rigs and they beat as close to the wind as seems to be possible

 

This leads me to another observation.

 

A sail driven boat moves forwards readily in response to a wind that approaches in the most suitable direction for the set of its rig. Winds in other directions do not produce a ready response. Then the sails flap to produce no drive and any sideways motion caused by wind drag on the hull and rig is impeded by the keel. The boat has a preferential direction of motion more or less in line with its heading. If a boat is sailing in highly disturbed wind it will sail in fits and starts in the direction that it is heading. If that direction is controlled by paddling, and not by sailing, progress in any direction is possible.

 

I would not be surprised to find that trees are planted in strategic places on the Broads to make the wind swirl and permit progress that would otherwise be impossible.

 

Now for models.

 

I think that most model boating takes place on sheltered ponds surrounded by trees and houses. To sail on such ponds with their swirling winds one must think of having a boat with a large unbalanced rudder to change course without sailing and a fin to limit leeway both when the boat actually sails and lee travel when it is simply being blown about. I think that this conclusion is inescapable. It tells me why Pearl is not suited to any lake but Maldon and perhaps Herne Bay when the wind is in the North.

 

Then the question that needs an answer is “is there any reason to provide sail control?” The answer must be that rudder control is all that is necessary and this decision gives another advantage. In light winds, even if they swirl, the sheeting cords that go through fairleads and so on put too much drag on the sails for them to set. Get rid of the fairleads and the sails will set in light winds. Time will be needed to get the best sail settings.

 

There are some consequences of this. If paddling is necessary a large rudder movement is needed. This makes the rudder very insensitive over its important middle. This makes sailing in a steady wind, where fine control of the rudder is needed, much more difficult. I suppose that one might use a dedicated Tx with switchable rates but I am not sure that the rudder control mechanics will be up to the job for the consequent tiny throw of the servo.

 

I have only just recognised this possibility of sailing in any direction in a swirling wind but one thought that follows is that models of, say, Victory may be possible but I would not be best pleased with the appearance of the model when on the move. To me the attraction is watching a model sail like the full-size in a steady wind with its sails full and drawing. I would rather sail once at Maldon in a steady wind than four times anywhere else in swirling winds.

 

June 2006

 

Since I wrote this I have made a model wherry with control of the clew and of the gaff. It sails so well that I am persuaded that setting the main with the clew at or about at the rail and the gaff out to give a very significant twist to the sail is probably the best that can be done.

 

Appendix 7 Setting up a Spektum computer controlled transmitter for controlling the sails of a bowsprit barge.

 

I am aware that this item will date quickly but the principles will not change. It costs nothing to add it to this site.

 

I think that most owners of a model barge would like to control it with two sticks, one for the rig and one for the rudder. This means that they want all the sails and the vangs to move together when one control stick is operated. In addition they would like the rig to have the best setting for the sails for every point of sailing. This is formidable if it is to be achieved with a two-channel radio set and an array of Y leads, pulleys and cleats. It is relatively easy to achieve using a computer controlled Tx. The programming is certainly less demanding than entering telephone numbers into a digital telephone.

 

So what needs to be controlled? The rig of a bowsprit sailing barge will comprise the mainsail together with the main topsail as one, the foresail, the mizzen and two jibs. Of these the foresail will follow full-size practice and not be controlled setting only at either end of the fore horse. This leaves four sails to be controlled by radio and of these four the mizzen can be regarded as a special case. That leaves three to be controlled by simple servos or winches hooked to the appropriate clews and each requires a channel. The vangs have to be controlled and these can be operated by one servo or winch using another channel. This is true for all computer Tx’s. Here I am interested in the Spektrum 7 channel outfit.

 

The Spektrum is designed primarily for aircraft use and, as a result its sticks are labelled for the appropriate aeroplane functions. These will be AIL, ELEV, RUDD, THRO, GEAR, FLAP and AUX2. We have to translate this to our own nomenclature for the sails of a barge.

 

Start by selecting a model memory. If it comes up for HELI you must change it to ACRO (see page 69) and then give it a name, for example, the name of the barge. Tie the Rx of your barge to this memory. Then it will not respond unless the Tx is set for the correct memory.

 

For sail control on the up/down of the left hand stick THRO becomes the master channel. We can let that stick movement control the mainsail. Plug the winch or servo that operates the mainsail into the THRO channel of the Rx. Moving the stick up or down will operate the main sail.

 

It is now essential the set up the mainsail servo or winch mechanically to operate with a travel of 100% because we want to feed this travel into the vangs and other sail servos and then set their travels independently. We do not want to start with some asymmetrical setting of the master channel. The mainsail will have the default setting of 100%.

 

Now we want the Tx to behave like a two channel radio. We have to reduce the response of all the channels except the AIL and the THRO to zero.

 

Scroll through the menu until you find (TRAVEL ADJUST). With the exception of the AIL channel and the THRO channel reduce all travels to zero. Then none of the switches or sticks will affect any servo except the stick for the rudder and the stick for the mainsail. The Tx is now a two-channel outfit.

 

Now we have to bring the other channels to life in response to the stick controlling the mainsail, ie, the THRO. We have to use the (PROG MIX) facility. Suppose that the servo for the vangs is plugged into ELEV channel. Scroll through to (PROG.MIX1) and press ADJUST to turn it on. Use INCREASE or DECREASE to change to the master channel THRO. Press SELECT to switch the cursor to the slave channel and change that to ELEV using INCREASE or DECREASE. This changes your transmitter so that both channels receive the input from the left hand up/down stick. Now select RATE and adjust the rate to give 10% each way. This small throw cannot jam up a servo or winch but does let you see that the mix is working and check that it is working in the required direction. Now both the main sail and the vangs should work together under the control of the master stick and operation of the ELEV stick should produce no result. If all is well the travel of the vangs can be set by going to MIX 1 and to the rate and set the %ages to get the right shape and travel of the mainsail as it sheets out.

 

This can be repeated for the two jib sail controls on mixes 2 and 3. Then the three sails and the vangs should all operate on one stick and of course be capable of fine adjustment using the transmitter.

 

Controlling the mizzen

This now leaves 3 mixes unused and this gives an opportunity to do some sophisticated control of the mizzen. Two servos are required They can be set up side by side and linked to the mizzen as shown.

 

If the mizzen left is on the rudder channel and is mixed with the mainsail by mixing THRO into AIL and the rate set to 50%, it can be made to move backwards when the main is sheeted out. In a similar way the mizzen right servo on the remaining channel can be mixed using another mix so that its arm also moves backwards when the mainsail is sheeted out. This sheets out the mizzen and is shown in 1. When the main, ie THRO is sheeted in both servos can be move forwards by giving them a 50% rate as shown in 2 to sheet the mizzen in. In this way both servos move forwards together or back together when the main is sheeted in or out. If the sheeting cords are set up correctly for length the mizzen can operate in conjunction with the other sails of the rig and help to drive the barge.

 

 Now we can make provision to set the mizzen to windward. This involves letting the aileron stick have an input to the servos. Go to THROW ADJUST and give the AIL stick 50% travel in both directions. The two 50% throws ensure that the servo cannot overrun. Operation of the aileron stick  will add 50% throw to the mizzen left but will not move the mizzen right. The last mix can be used to mix AIL into the mizzen right channel which must be set to either +50% either way or –50 % either way as required. Then the mizzen can drive the barge in normal sailing and can be set to windward during a tack. In addition the right hand stick can move to left or right to set the mizzen independently of the main.

 

The mizzen can now be set to windward to help the barge tack. Normally the barge would tack from beating close to the wind. One would initiate the tack by bearing away a little to increase speed and then turn to rudder to make the head turn to windward. The mizzen can be set to windward using the AIL stick to help with the tack by weather-cocking.  It adds to the authenticity of the model as a scale model.

 

I am beginning to think that the balance of the barge can be improved by using this facility to alter the angle of the mizzen independently of the main when needed. It may be more important than being able to set the mizzen to windward for a tack. But I need a steady wind to find out for certain.

 

Appendix 8 Pictures of barges under way.

I have found it to be difficult to get good pictures of Thames sailing barges when they are actually sailing. If part of the object of making the scale model is to make it sail like the full size this is presents something of a problem.

 

I think that most of the pictures that we have show the Thames sailing barge doing something exceptional probably because the sailing barge was just a part of the landscape and not of special interest. Furthermore it is hard to see how photographs of barges under way could have been taken other than from another boat. Add to that the fact that we need photographs of barges as they were sailed and the shortage of pictures is inevitable.

 

I was talking to Cyril Bagshaw about this and he volunteered the photographs in the block of four given below. I am grateful to him. In the first three pictures the barges are beating to windward and the mainsail clews are in line with the rails. In the fourth picture the barge is also beating but the clew has been let out well beyond the rail and the camber in the foot of the sail is very evident.

 

 

 

 

 

 

 

 

 

 

The group of two pictures are from Ken Fletcher who followed a barge match in another barge.