Articles are dated and revision noted.  All articles are subject to change as new data becomes available!

The author sails with the local IOM club in the Northern Beaches of Sydney, Australia, winning some races but often not.  Thus the IOM Class is often used for experiments in this website

The author is definitely not interested in any commercial activity concerning radio sailing.

The site and its articles are copyright but readers are free to use or quote anything from the site PROVIDED DUE RECOGNITION is made of the source.

Thanks to Lester Gilbert for his pioneering site www.onemetre.net that encouraged me to do this one.

1.0    Introduction
This article describes observations of wind gradient and wind speed plus direction measurements at Lake Julianne in North Queensland, Australia.  This is the home of the Whitsunday Model Boat Club, situated in the tropics at approximately S 20 deg. 20 min, E 148deg. 36 min.  It is approximately 9 kilometres in from the coast and is approximately 200 metres east west and 170 metres north south.  To the south and west the land is flat with sugar cane fields, while to the north and east are gently rolling hills with dwellings.  

The observations were made in winter when the most common breeze is a SE trade wind, however by chance most of the observations were in a NE wind. The photograph above is not on any of the days of observations!

The observations were as follows:

•    Gradient measurements from the pontoon in NNE wind on 19th September 2009.

•    Wind speed measurements at 2 second intervals for about 10 minutes in a NE sea breeze on 7th September at a point about where the boats are in the picture. Air temperature 24 degrees Celcius.

•    Observation of wind speed while sailing an IOM with an A rig around the same point as above in a SE trade wind.

•    Observations of wind speed and direction at 2 second intervals for about 15 minutes on 14th September at the same point as above in a NE sea breeze. Air temperature 25 degrees Celcius.

2.0    Measurement of wind gradient
The method adopted was to measure the wind speed at 2, 1.5 1.0 and 0.5 metres and 50mm above the water then move onto the next height.  This gave a large scatter of wind speeds for each height: not surprising given the variability in the later readings  described below.  Reasonable gradients were, however, obtained by averaging a large number of readings.  It is interesting to note that significant wind speeds were measured even as low as 50mm above the water surface.  The results are given in Figure 2.  The sharp increase near the top is thought to be due to the proximity of the bank behind the measurement point.

For a boat sailing at 45 degrees to the true wind at 2 knots the measured gradient requires a twist over the height of an IOM A rig mainsail of just under 9 degrees.

3.0    Observations 10th September 2013
On this day the wind speed at the average height of an IOM sail plan (about 670mm above the water) ranged from 6 to 11 knots from the SE.  The measured wind speed showed similar rapid changes to that noted below, however the behaviour of the IOM with an A rig  sailing around the observation point did not show any evidence of this second to second variability, proceeding smoothly both up and downwind except in the odd major gust.   This shows that the effect of lateral inertia and damping by the water acting on the keel is substantial. 

To windward the gunwale came level with the general lake surface at a wind speed of 7 knots, but because of the trough between the bow and stern waves the gunwale was clear of the water. The boat stayed under control to windward throughout, downwind however, control problems and nosediving occurred once the wind reached the 11 knot range.

4.0    Observations 7th and 14th September 2013
The wind on both these occasions was a NE sea breeze and the readings were taken during a break in the racing.  None of the skippers though the wind was unusually gusty or shifty though it was recognised that in some races a large wind shift did occur, and an occasional nose dive occurred.  Once again the boats did not reflect the extreme variability in wind strength that was observed, proceeding fairly steadily without sudden changes to the heel angle except for a few major gusts.  The results are presented in Figure 3A for the 7th September and Figure 3B for the 14th September.

The wind direction results pose some interesting problems.  The normal windward leg lasts about 3 minutes.  If the race began at time zero in Figure 3B the boats would experience a general lift to starboard for the whole leg (tacking on the “noise” would be unlikely to be fruitful): the correct tactic would therefore be to tack onto port at the start and go to starboard about halfway up.  This is like the conditions in the Kiel Olympics which caused consternation for people whose only experience was in oscillating winds and required a counter intuitive tack onto the knocking tack at the start.  On the other hand if the leg began at about the 3 minute mark then it would be correct to start on the lifting tack.  How one could ever tell the difference I cannot say, but it would be useful!

Analyses showed that the gusts could either veer or back with no preference for either direction.  Likewise there was no connection between wind direction and wind speed.

What statistics to extract is still being developed but Table 1 shows those extracted to date. Some of the items deserve explanation.  

The Gust/ Lull ratio is the upper wind speed in the gust divided by the wind speed just as the gust begins.  The gust onslaught rate is the speed at which the gust gathers strength.  Depending on the magnitude of these indicators, different tactics need to be adopted to parry or use the gusts.  In our NS14 dinghies we had a range of options from easing sheets, swinging harder, or even changing the flexibility of the rig  In model yachts the choices are more limited, once the boat is on the water the choice is whether to ease the sheets or let the boat cope.  The wind change over 2 seconds gives an indication of how gusty things are and the wind direction change over 2 seconds shows how shifty the wind is.

Hopefully with data from other locations these will give an indication of the nature of each course and provide guidance on handling the conditions.

Col Thorne Jan 2014

•    As David Hollom suggested in the Seattle newsletter, the presence of chines aft allows for a narrower band of separation at the stern when heeled and this is likely to reduce drag. Of course this only works when the boat is heeling enough.
•    Depending on the design, it seems a reasonable speculation that the wetted surface and the drag when heeled well over may be significantly less with a correctly placed chine and tumblehome than for a slab sided or flared cross section.  Hopefully I will get round to looking at this in a more quantitative manner sometime!
•    In dinghies I designed two round bilge boats.  These were very good in waves and the first one got first and second in the championships that year.  The second round bilge one had hollow waterlines forward and a very low wetted surface, and was faster than the first. It proved very quick upwind, especially in waves, and very quick though the transition from displacement to planning but it fell short of the chine boats in flat out planning, though it was pre-eminent for river sailing where flat out planning was rare.  In hindsight it would have been better to have a chine over the last half to one third of the hull to allow the water to break away more cleanly from the aft sections.  This would have provided a bit more dynamic lift and also would have reduced the wetted area by stopping water from running up the flared sides aft.   Frank and Julian Bethwaite did some experiments on how sharp the chines should be, and came up with a 2% gain by going from pencil round to a sharp chine aft.
•    Several designers have said that they believe the topside on the lee side projects a flat surface and that helps reduce leeway.  Personally, this seems unlikely to be a significant effect but the truth is that I have no data.
•    There is another rather exotic possibility.  I have noticed that when pressed to windward near hull speed, some IOM designs develop an exaggerated trough in the bow to stern wave and I would speculate that this may be due to the low pressure area on the windward side of the keel pulling the trough downward.  If the chine allows the top of the keel to be a bit lower down from the surface, this might (?) be avoided.

It is interesting to speculate what the next jump will be. There are designs out there like the D2 IOM from New Zealand and some others in development here in Australia that address wave impact.  The ones I have seen are going down the route of the bow shapes on the current crop of America’s Cup multihulls, the idea being that the bow can go into the wave and emerge without much effort.  There will be nosediving and other issues with this, but I expect you clever designers will come up with something workable.

Colin Thorne July 2013

I first came across the idea of a set of regimes dependant on wind strength in a book by Walker written in the 1960’s.  He based the regime limits on the relative power of the sails to produce side force to windward as against the ability of the rudder and keel to resist it.  It was a difficult description but set the stage.  The next serious look at splitting the range of wind strengths was Frank Bethwaite’s wonderful “Wings on Water” series.  He used a combination of wind speeds and boat stability to define his regimes. 

In model boats both these ideas come into play. 

The blue curve in the figure shows the approximate boat speed attained by an IOM with an A rig in flattish water versus the wind speed at the centre of area of the sails: note that up on the bank the wind speed would be 50% to 100% higher.

Now the sail side force is proportional to the square of the wind speed and the maximum force that can be generated by the keel and rudder to resist this side force is proportional to the square of the boat speed. Thus the ratio of (wind speed speed/boatspeed) squared indicates the relative “power” of the sails as compared to that of the keel and rudder. The pink curve shows this ratio, so where this is small, the keel and rudder are more powerful than the sails and vice versa. From the shape of the curve, it is evident that different factors are at work at different wind strengths.

In the drift regime, both the wind strength and the boat speed is low.  Also in these conditions the gust to average wind ratios are high (a half a knot to 1.5 knots is a 9 fold force increase!).  Lift/drag ratios are also poor, which results in the boat speed being relatively small compared to the wind speed.  What this means is that it is easy to overload the keel and rudder because they are moving very slowly through the water so they have little capacity to generate side force and when a gust hits, acceleration will be sluggish.  Traditional wisdom is that in this regime flat, heavily twisted sails are what is required in what are usually glassy conditions.  The reason is a perception that a flat plate is the best shape in very light airs (think butterflies), and in such conditions there is usually a steep change in wind speed with height starting from zero at the water surface. It is possible, however, for such light winds to be turbulent if the pond surrounds disturb the airflow and this may ruffle the water surface: in this instance less twist and fuller sails can be better.  In either case sailing for boat speed rather than pointing is the way to go.

In the acceleration regime all this changes.   The acceleration regime begins as the wind becomes stronger: an indication is the development of a small stern wave.  Lift to drag ratios increase and the boat speed climbs relative to the wind speed.  In this regime too, the gust to average wind ratio tends to decrease.  In this regime the hydrofoils can easily develop the necessary side force so the increase in side force as a gust hits can be accommodated without an excessive leeway.  As the boat accelerates the leeway will decrease again.  In this regime, most boats will sail themselves. Even boats with the keel a bit too far aft and carrying lee helm or boats with small keels may sail reasonably because the high side force capacity of the keel will mask these defects.   In this regime the keel and rudder can easily cope with any side forces from the sails, so the boat can point high and have powerful sails. The top of this regime is reached when the boat is approaching hull speed, or rather when any increase in speed requires a disproportionate increase in driving force, coupled with excess heeling.

The feathering regime begins after the acceleration regime ends. When this happens, 
gusts can produce major increases in side force, heeling the boat and loading up the hydrofoils with no hope of increased boat speed to help compensate.  The result is increased leeway and weather helm.  The situation is exacerbated with waves because they can easily double the hull resistance, slowing the boat and further reducing the ability of the hydrofoils to produce side force.  This regime will show a marked decrease in performance of a boat with an undersized keel, and if the boat still carries lee helm in these conditions it will suffer severely.  The boat may well become sluggish, staggering in the gusts and generally slowing. In this regime, sails need to be flattened and high pointing is not feasible. 

This, then is the general idea of regimes.  The boundaries for B and C Rigs are determined by similar factors, though the wind speeds and boat speeds will differ. Of course subdivisions can and have been made based on such things as gust onslaught rates, gust/lull ratios, waves etc, especially their impact on boat handling, but all that is beyond this short article.

Colin Thorne
September 2009 rev June 2013

Towing in a swimming pool has the problem of getting to a steady state quickly.  There is also the problem of measuring quite small forces accurately.  The setup below is one solution to the latter problem.

The arrangement shown will work provided the balance (or scales) has graduations down to 1 gram.  Then it will be mostly correct to about 2 grams.  The beauty of this setup is that it is not “springy” and hence will not contribute to oscillations in the boat movement that are inescapable in a springy setup.  The towline needs to be as stiff and light as possible, a strechy one will also cause oscillations.

The drag from the boat will try to lift the weight up and reduce the balance reading by the same amount as the drag.  Thus if the balance originally read 1002 g and reduced to 902 g the drag would be 100g.

The rollers can be bits of round disused mast section.  Probably about 4 would be OK and the smooth flat surfaces can be laminex over a flat board or smooth ceramic tiles. 

The towing module needs to be self contained and the pulley must be fixed relative to the Balance.

© Colin Thorne March 2013

In IOM’s it has become commonplace to pre-bend a top portion of mast forward so that when the masthead is pulled aft with the backstay to suit the cut of the sail, a greater forestay tension is produced than could be done otherwise

Although the primary job of the pre-bend is to increase the forestay tension, it does, together with the check-stay/mast ram and spreader set up, set the sailing shape of the mast and hence the sail shapes. 

For some time, a pre-bend offset (Δ in the Geek Box) of around 40mm was used, however both this, and the length of mast section prebent, has been progressively reduced by most IOM sail makers. The result is that many club sailors find that when they but new or near new sails these will not fit happily on their old rigs.  If the maker of the sails will reveal what prebent is required, well and good.  It is just a matter of rebending the mast.to that specification.

Changing the bend in the mast is not difficult, the most important thing to remember is that the bend MUST be only fore and aft.  I put the spreaders in so that I can make sure of this.  Some just use their hands, but more often the rebending is done over a round object.  A car tire is good and often used: you place the mast on the flattest part in the middle, check the spreaders are horizontal and the gooseneck vertical then press down either side of the point of contact.  Start off gently until you get used to how it all works.  If this all looks too uncontrolled for you, a jig can be made like Figure 1.

On the other hand, if this information is not available, all is not lost!  As the backstay is tensioned to pull the mast to its sailing shape the mast goes through some rather odd S bends, and the final shape is determined by how far it is pulled and by the mast ram setting, the spreader length and angle, and the shroud tension.  Of course the final mast shape affects the sail shape.  If the shape is not right then it is possible to vary the sail fullness throughout most of the sail by changing the shape of the pre-bend.  This depends on two factors:

1. Parts of the mast that are bent to a tighter radius on the pre-bend tend to become relatively straight once the backstay is tensioned.  Conversely, parts of the mast that are relatively straight become more curved as the backstay is tightened. 2) The sail is flattened where the mast takes up a tighter radius with the backstay on.  Conversely the sail will become fuller in sections of the mast that become relatively straight with the backstay on.

This can be useful.  My first attempt at a B rig had too much fullness near the head and, because the mast was a rather floppy tube section, tended to get a nasty over-flat shape near the spreaders with creases to the clew.  No amount of mast ram stopped this and initially resort was had to bending the spreaders forwards, which still left the head too full. 

The pre-bend was changed by straightening the top section and putting in a tighter bend just above the spreaders.  The total amount of pre-bend as measured by the tip deflection from the line of the mast below the spreaders (25mm) was kept the same. This fixed all the problems and the spreaders could be put back to a more normal angle.

As another example, I recently bought a second hand boat with a very nice mainsail but with a mast that was clearly wrong.  The sail was far too full in the mid-height region.  On inspection there was considerable pre-bend in this area, so that when the backstay was tensioned this part of the mast was actually bend slightly backwards!  By removing the pre-bend from this mid-height area, the whole sail looked very nice when the backstay was tensioned, though it took a couple of goes before it was right.

Some people like to set templates for a uniform radius for prebends.  For the old 40mm at the tip, a full height bend from spreaders to the upper point implies a radius of 13.525m.  Scott Berry says for the Blowfly the bend should be only from the spreaders to the band above the forestay point, which requires a radius of 12.9m. These calculations allow for the difference between H and L in the Geek Box. The difference is small and ignoring this difference changes the offsets by less than 0.1mm. 

Most modern rigs use smaller prebends.  If you are keen, the Geek Box shows how to calculate the offsets for a template for any given radius or start point.

Colin Thorne April 2010, rev May 2013

1. 0 Where to put the keel?

It should be clear from the above that the primary need is to have the keel in the correct position relative to the rig. In most development classes of model boats, the ballast is in a bulb at the base of a deep narrow keel and the ballast ratio is upwards of 60%.  Given this, the bulb has to end up in a fore and aft position close to that of the centre of buoyancy.  Of course, one can juggle with the heavier items like the winch and battery but other constrains make these changes relatively small. I have seen a 36R, which has a wider and much shorter keel than the IOMs or Marbleheads or Ten Raters, where the bulb extended well forward of the keel, and the keel and rig were well back from the usual position. Reportedly it sailed quite well! Having said that, there is interaction between the keel and hull and this is more a matter of trial and adjustment within the range possible.  With long thin keels there can be problems with torsional distortion of the keel. (See article under “Keels etc.").

In the usual development classes of model yachts in Australia, the keel position is largely dictated by weight balance.  The rig then has to be positioned to suit the keel, and class rules.  Most new designs need some post launch adjustment to get it all correct. It is theoretically possible to work out the desirable keel/rig placement using diagrams like those in this article the uncertainties of the magnitude and line of action of the hull side force poses the main problem. This was the method I adopted for my Trailer Yacht designs and it worked well. Most designers simply look at successful boats and then adjust from there. 

If you are determined to experiment (an attitude I applaud!) then remember that a little too big a keel or rudder is much better than a little too small, and a keel a little too far forward is a lot better than one a bit too far aft.   Of course a perfectly sized and placed keel/rudder combination is best of all but this is likely to be perfect in only one set of conditions anyway!

2.0 How much things effect balance

The following table gives approximate movements of the intersection of the hydro and wind forces with the hull center-line for various adjustments.  These are approximate but do help to understand what does what and how much. The figures for changes to the sail forces are fairly straightforward, but those for the keel and rudder forces need some explanation because they involve some assumptions.  For the purposes of the calculations it was assumed that the hull contributed 5% of the side force and that this acted at 300mm from the bow.  It was also assumed that the rudder had an area 40% of that of the keel. 

It was also assumed that, for the base case, the boat was set up to sail to windward in steady conditions with the rudder on the centreline.  Because of the downwash from the keel the angle of incidence of the water flow on the rudder is reduced to just over half of the leeway angle, and because the rudder has a lower aspect ratio and operates at a lower Reynold’s number, it produces less side force pro rata than the keel.  As a rough guide the rudder will then produce a side force per unit area about half that of the keel with the rudder central and this is what was assumed for the base case.

Increased side force means more leeway, at least until the boat accelerates to increase the lift from keel and rudder.  Of course, if the boat is already at hull speed, very little increase in speed is likely but all that is too esoteric for this article.  Most of these effects, i.e. increased weather helm and increased leeway, will slow the boat, or at least reduce any acceleration.  The techniques to handle gusts depend on the boat type.  Clearly in a displacement hull already at hull speed the idea is to minimise the increase in side force while luffing to take advantage of the swing in the apparent wind.  On the other hand in light dinghy that can plane to windward a different approach would be adopted.

The increased load on the sails and rigging will also cause them to change shape to some degree.  Ideally, the upper sections of the sails would flatten and feather off, easing the upper leech.  If you put a properly set up Soling one metre on its cradle in a proper angle to the breeze and watch the sails you will see this effect, though to my view the response is a bit too great.  However, not all such changes in the sails are beneficial.  A common change in soft sails is that the draft blows aft and the leech hooks.  This results in an increase in the wind drag force and a decrease in the wind lift force (and hence a reduced driving force).  In addition the force of the wind on the sail also moves aft and this increases weather helm, while the increased side force leads to excessive heeling.

Some other effects are a bit more subtle but equally or more damaging.  As an example I can cite my last NS14 (a 14 foot development class that formed the basis for the International Tasar).  All my previous ones had the mast stepped on the centreboard case with chocks at deck level (rather like an IOM), with a powerful boom vang that was taken onto the hull just aft of the base of the mast.  This vang acted as a backstay as well as a vang.  On the river, the gusts hit too fast even for an NS14 and we commonly shed a foot or so of mainsheet for a second or two to take out the sting out of the gust and to help acceleration while we luffed gently.  When I did this in my new boat it staggered rather than accelerating like the previous boats did.  I knew the main was doing the right things, but careful observation showed that the forestay sagged at each gust and the sheeting geometry meant the leech actually tightened.  This occurred because in the new rig the vang came to the base of the mast at deck level and did not have any “backstay” effect so when the main was eased all the tension went from the forestay.  The problem was reversed by using dramatically increased side stay tensions and adjusting the jib sheeting.

 If the backstay of my Soling one metre is over tensioned, something of the same sort occurs, though with a different mechanism.  In a more usual model yacht with a topping lift on the jib boom and if the pivot point is too far aft, I can imagine a similar problem if the topping lift can stretch because, when the gust hits, both topping lift and forestay will be loaded and if the topping lift can stretch, the see-saw action of the jib boom will slacken the forestay while the cloth in the leech could be tensioned.

Another hypothetical adverse behaviour could arise in a rig where the back stay was set up so tightly that the mast was on the point of buckling.  The extra load on the forestay and could then conceivably buckle the mast which would shorten it and allow the mast head to move forward and hence allow the forestay to sag.

In a lull the reverse to all this occurs.  This can have its own problems.  An example is the Soling one metre: if it is set up to look right in a wind that heels the boat 20 degrees or so, when a lull comes, the leach of the headsail springs back too far and chokes off the airflow. I am sure that experienced model yachties can think of many more examples!

Colin Thorne

July 2008

A couple of years ago I designed and built a test rig so that the behaviour of sails under load might be studied.  The rig heels the same as an average IOM under wind load and so it is possible to look at the sail shape as a function of the heel angle.

The rig is shown in the Figure 1 and cameras mounted on the rig provide photos like Figure 2 that, using some free software from UK Sails, can be analysed to show the camber and twist at various heights up the sail.  One such study looked at the behaviour of a 2003 IOM A rig headsail.

The mast was pre-bent to the then usual 40mm deflection as in the Sails Etc drawings, and the material was the lighter grade film.  The swivel was 75mm behind the headsail luff.  The bulk of the readings were taken with a backstay tension of about 20N, which gave an unloaded camber of 7.5% at the top batten when measured inside with the boat on its side. 

Some readings were taken for other backstay tensions, though the wind became quite gusty and the rig hardly stayed at one angle of heel for more than a second or two, making measurement difficult and the data showed a large scatter as a result.  It became clear that this was because the rig has less inertia and damping then an actual boat sailing, because boats sailing were not changing heel angle anywhere near as rapidly, though they must have been experiencing similar wind variations.  This just emphasised how much micro structure is in the wind at model yacht level.

The resulting change of camber with heel angle is shown in Figure 3 for the top stripe and Figure 4 for the centre stripe.  The top stripe result shows a progressive increase in camber with increasing heel angle, exactly the opposite of what is needed.  The angle of attack did not change, indicating that the leech and luff fell off athwartships at about the same rate.

The centre stripe (Figure 4) shows a similar result to that for the mainsail in Part 1 in that after a change at low heel angles, the camber stayed roughly the same as the heel angle increased.  The twist did not change substantially over the whole range.

This is surprising result because it would be expected that the sag in the headsail would have to increase with increasing wind load.  What actually seemed to happen was that the extra load was taken by the cloth near the forestay with the result that the camber moved forward with a tight curvature near the forestay.  At least these results are not as bad as those described by Lester Gilbert in the wind tunnel tests.

An attempt was made also to look at the camber for different backstay tensions at 15 degrees and 30 degrees angle of heel for both the top batten and the mid height of the headsail. The results are shown in Figure 5.  It is interesting that the change in camber with backstay tension seems to be similar for the different heel angles.  It would seem more likely that the difference would be greater as the boat heels more. The answer seems to lie in the tension taken by the cloth near the forestay. 

There is a trend too for less backstay tension at setup with attendant less forestay tension.  Figure 5 suggests that with a 1kg backstay tension sails need to be cut about 3% flatter at least to come out the same at those for higher backstay tensions.

Col Thorne May 2007

Part 2 Effect of Changing Wind Direction and Wind Speed

1.0 What happens when the wind changes direction?

Most racing model yachts, including my Soling 1metre and my IOM are able to sail to windward and follow moderate wind shifts without the helmsman doing anything for some period of time:    To understand how this can occur we need to look at what happens if the wind changes direction.

Figure 2 shows the water and wind forces during a period of steady wind, shown by the full lines on the figure. The water and wind forces balance and line up, both forces intersecting the centre line at point C. If there is a lift in the wind, i.e. the wind backs in the diagram, the wind force will move with it from the original position AB to some position like AB’(shown as a dotted line).  Now the line of action of this new force goes behind the existing water force, intersecting the centre line at C’ creating a moment that will turn the boat up into the wind, decreasing the apparent wind angle. This will be made a bit more dramatic since the larger angle of attack of the wind on the sails will increase the wind force, heeling the boat. (A discussion of the effect of heeling is given in the next section).  When the apparent wind angle reaches the old value the forces line up again and the boat will steady onto a new course.

If the wind heads the boat, the reverse occurs.  The wind force will rotate the other way and the line of the force will move ahead of the water force creating a moment that turns the boat away from the wind.

One important matter that comes out of this is that if the sails are too far out or with both leeches off too twisted, the boat may sail nicely but too far off the wind.  The reverse is also true if the leeches are too tight.  Thus just because the boat is well balanced does not mean it is at its fastest.

Of course if the wind heads too much too quickly the boat may end up in irons or on the other tack!  There are some things that can mess up this automatic response.  The primary one is wave action but the rotational inertia and water resistance to turning of the boat may either cause a response that is either too slow or too fast.

2.0 What happens when the wind strength increases?

Everyone who has raced sailing boats has experienced the euphoria of finding themselves keeping up with the hotshots, then a few gusts come through and the euphoria dissolves as the hotshots suddenly gain many boat lengths.  It is an illustration yet again that it is not steady state straight line speed that separates the hotshots from the rest so much as the ability to handle unsteady conditions.  Now the techniques for handling gusts and lulls vary from venue to venue and class to class and the details are arcane to those with the particular experience required. However, there are a few common principles and that is all we can cover here. 

The full lines in Figure 3 show the steady wind situation, with wind and water forces equal and opposite.  When a gust comes, even if the true wind direction does not change, this equilibrium is disturbed in several ways.

  As the wind force increases it causes the boat to heel.  This has the effect not only of increasing the wind force but also of moving its point of application from the initial point A to a point A’ further outboard so that the line of its force intersects the centre line further aft at point C’ instead of C.  The increased true wind will also mean a lift in the apparent wind direction that adds to the effects above.  The heeling of the boat also means that the line of action of the water drag forces change, the hull drag goes to leeward while the keel and rudder drags go to windward. However, the net effect of these movements on the line of action of the resultant of the water drag forces is small relative to that of the wind forces. Several things follow from these effects.

The wind force tries to turn the boat to windward and more weather helm is needed to bring the water force back into line with the new wind force and this effect is probably exacerbated by the hull side force moving forward as the boat heels, though this depends on the hull shape. If the helm is not touched, the boat will head closer to wind and this will tend to decrease the side force and restore the equilibrium as it did for a wind shift without any wind strength change.

Colin Thorne

July 2008

Part 1 The Concept 

1.0 Introduction

The winds we sail in are unsteady, and vary not only from hour to hour and minute to minute but second by second.  These changes occur too quickly for a crew to reconfigure the sail settings, and an ideal rig will alter automatically to suit these rapid changes.  There are two different aspects to an “automatic rig.”

(a)  Below the design wind flexibility to adjust to micro and macro wind speed changes in such a way as to maximise acceleration and minimise increases in side force.  In simple terms this means the upper leech falls off a controlled amount due to increased wind strength and apparent wind angle in gusts and vice versa.

(b) At and above the design wind to flatten and feather the sail, especially the upper parts, to minimise drag and heeling moment.  Flapping must be avoided.

The design wind is fairly well set in dinghies where the maximum heeling moment occurs close to no heel, but is rather less well defined for non-canting keel yachts and can vary from around 15 degrees for light flat boats up to more for thinner ones.  For an IOM it is probably about 25 to 30 degrees or so.

The first time I experienced this was in the NS14 class: the difference was astonishing both for the ease of sailing but also the speed advantage, with gains of 4 to 5 minutes round a 90 minute course, even though our thinking was not this far advanced.  

2.0 How is it done?

The primary tools that designers have to achieve this automatic response are:

1)  Mast flexibility and its distribution.

2) Rigging flexibility

3) Sail cloth stiffness.

4) Plan shape of the sail (large roaches and stiff battens up high help, especially with (a) above).

5) Sail cloth stiffness

6) Batten stiffness (full length battens)

7) Sail structure (if camber is put in by seam taper rather than by luff round the leech stands up better.)

8) Prestress of the sail, rigging tension and mast bend at setup.

In all cases the driver for auto response is the wind induced changes in the leech tension of headsail and main.  To evoke (b) above this needs some form of prestress so that the sudden change occurs at a predetermined wind speed.  The elastic response (a) should not be overdone otherwise too much power can be shed.

Rigs that behave like this are now used in many dinghy and even yacht classes. Bethwaite (1996) describes the system used in the B18 class: this had a mast that was shaped and rigged to be rigid below the hounds and relatively floppy above.  By using a very high luff tension on the main, the mast was prebent so when the wind increased to the design wind strength, the increased luff tension became enough to make mast bent further and the sail feathered.

There are now numerical modelling programs used for large yacht design, and the buzz word for this is aero-elasticity, however for one set of runs you could buy a fleet of IOMs and I am not sure if the algorithms are right for air and water flow at the scale of model yachts, so approximate calculation and testing are a more realistic bet.

With this background I set out to find how model yacht rigs stack up, especially the IOM class.

Some class rules mean that the choice is restricted. For instance in the IOM class the mast has to be uniform for the full height and the sail plan shape is tightly controlled.  In model yachts there is not a lot of choice for sail materials and most are rather stiff relative to full size sails.

As a first look at an IOM rig I used an early apparatus as in the picture, and did a set of tests on a B rig  that could later be quantified.  Results for twist in the mainsail at the upper and middle battens are shown in the chart for an apparent wind angle of about 40 degrees and a backstay tension of 25N.

The lower axis is the amount of heel caused by the wind speed when the readings were taken.  The numbers at zero heel were obtained inside with the mast horizontal so only the weight of the sail cloth pulled the sail down.  It can be seen that initially the twist changed rapidly with heel angle, with this process certainly completed by a heeled angle of 15 degrees, and maybe before that.  This change is probably due to the different distribution of wind pressure, which tends toward a quadrilateral shape across the chord with a maximum near the mast, as compared to the cloth weight which is uniform across the chord.

Beyond 15 degrees of heel the twist changed only a little, less than one might hope for in an automatic rig.  The camber hardly changed at all over this range and the sail stubbornly held its camber in spite of increasing wind pressure and clearly showed no ability to adjust to wind change without intervention, rather like the old telegraph pole rigs we had on dinghies before the mid 1960’s when we began to work on “automatic” rigs.

This does seem to be changing in the IOM Class, but is not widespread.

The following parts in this series describe some tests and analyses on this topic.

Reference

Bethwaite (1996) High Performance Sailing

Colin Thorne March 2013