| 
Windvane self steering gears
Of the many delights of sailing, I believe steering by hand is the one that palls the fastest. Yes, it can be exhilarating at times, even satisfying when you’re otherwise bored, but it becomes totally insufferable when embarked on anything more than a short passage. To be able to leave the helm and engage in other more constructive or entertaining activities is an immense boon to the offshore sailor.
The solution for many is an electric automatic pilot. These are relatively cheap to buy and easy to operate. You simply switch them on and, by virtue of sophisticated computational algorithms, they will more or less do their duty. Unfortunately, such convenience is not without its downside. They are also quite noisy, in my experience somewhat unreliable, and are inherently wasteful with a sailboat’s meagre store of electricity.
On the other hand, wind driven self steering gears are silent, seemingly indomitable in their reliability, and consume no electricity whatsoever. Hardly surprisingly, they are the first choice for ocean voyagers and, in my opinion, would also serve coastal sailors much better than most would appreciate. In the debit column we must acknowledge their higher initial costs and the greater responsibility required of the crew to have their sails set and trimmed properly, a call for good seamanship, no less, something we should be practicing anyway.
But this isn’t a head-to-head between electronics and mechanics. There’s no competition. Windvane gears won’t work in the absence of wind and there’s no shortage of electricity when your engine is running. In common with many offshore yachts, mine carries both: the tiller pilot doing duty in the very dreariest of steering conditions – a flat calm – and the windvane standing watch when under sail.
Electronic autopilots are usually controlled by compasses, whereas windvane gears sense the wind direction and keep the boat’s head at an angle relative to it. More correctly, they sense the apparent wind, which is the wind as experienced aboard the boat – a combination of the true wind and the forward progress of the boat itself. Between the actions of the two types lies an important distinction, each with inherent merits and traps. The compass-controlled gear will deliver the kind of straight line course you can mark on your chart. This makes for very predictable navigation. Not so the windvane which will duck and weave with every variation in wind direction and strength. On the face of it, this might appear a serious flaw until you remember that the sails are trimmed according to the wind and that it’s actually beneficial to respond to its inconsistencies if you’re going to get the best out of them. Indeed, because it never relaxes its concentration, a windvane will often take a boat to windward more surely than will a human hand.
Direct action gears
Much of the development of windvane steering arose from singlehanded sailing. In 1936 the French painter, Marin-Marie used a windvane to cross the Atlantic in his motor yacht Arielle. The first recorded use on a sailing boat was in 1955 when Ian Major also crossed the Atlantic in Buttercup. 1960 saw the first Observer Singlehanded Transatlantic Race (OSTAR). Amongst the five competitors were Colonel HG (Blondie) Hasler war hero, instigator of the race and later credited with being one of the inventors of ‘modern’ self steering gears and Francis Chichester with his 39 footer Gipsy Moth III, then thought to be about the largest yacht one man could handle.
To steer Gypsy Moth III while he rested, Chichester fitted a rotating (and reefable) mizzen, like a gigantic weathercock, connected directly to the steering. It was operated by putting the boat on course, allowing the vane to feather into the wind, then tying off the tiller lines. With uncharacteristic whimsy, he called it Miranda. Despite her size, Miranda was a mediocre performer, producing too little power and lacking the sensitivity her femininity suggests. The friction in the whole system can only be imagined, but that wasn’t the only problem, as we shall touch on in the next section. For now let’s accept that the output from any direct action gear is unlikely to be enough to overcome the loads found in most primary steering systems.
The search for sensitivity
The main problem Francis Chichester encountered with Miranda is inherent in all vanes rotating about a vertical axis. Since the power output from such a vane is dependent on its angle of attack with the wind, it follows that the boat must stray well off course before any appreciable angle is achieved. Put another way, if, say, the head falls to leeward 5º, the vane will present itself to the wind at only 5º and such a slight output can easily be lost to friction and the inevitable slop in the various linkages and lines. Then there’s the shape of the vane itself. Although an efficient aerofoil will generate lift at 5º, a flat sheet of plywood (let alone Miranda’s flapping sail) would scarcely notice it. Only when the boat veers grossly off course will the gear respond. Some means of obtaining a vane output greater than the yaw angle was needed.
The first answer was to pivot the vane horizontally (left). Now, if the wind pressed even slightly on one side, the vane will be pushed over. With this arrangement, the vane’s rotational output is no longer limited by the yaw angle. It becomes both stronger and more emphatic.
Indeed, in practice, horizontal axis vanes proved too powerful and twitchy. In strong winds the vane would slam from side to side, inducing horrendous oversteer. To overcome this, a modern gear has its axis inclined away from the wind at about 15-20º. This has the effect of progressively feathering the vane as it’s pressed over (see left), thereby dampening the output at the extremes of vane rotation and reducing the associated course oscillations. With an inclined axis of 20º the vane’s angle of attack becomes zero at 30º of rotation. Incidentally, heel angle must be added to the axis angle, so an inclined axis vane becomes significantly less powerful if the boat is hard pressed.
Indirect action
Steering calls for strength. There’s no doubt about that. Boats live on the interface between two turbulent fluids – air and water – which continuously buffet and toss them about. Even on the best trimmed boat, the steering loads can be high – certainly more than could be consistently overcome by the puny output from a windvane acting alone. To achieve efficient self-steering, a source of greater power is needed. Fortunately, there’s one to hand: the flow of water past the hull.
Let’s go back to that 1960 OSTAR. While Francis Chichester was flirting with the lacklustre Miranda, Blondie Hasler on his turtle-decked Folkboat, Jester, had adopted a trick from aircraft design. On the trailing edge of Jester’s transom-hung rudder was a small trim tab, driven by a windvane that Hasler could adjust from the midships hatch. Whenever the boat strayed off course, the vane turned the tab and the waterflow acting upon it pushed the rudder blade over to make the appropriate correction (right). This could be described as having a tiny auxiliary rudder that steers the main rudder, which in turn steers the boat. In this context it’s the simplest form of servo effect, where the main mechanism – the hydrodynamic lift on the trim tab – develops a much greater force than the force communicated to it.
This process of amplification means that the windvane can be quite small though, interestingly, Hasler clung to vertical axis vanes in all of his later, and considerably more sophisticated, designs. A disadvantage of trim tab gears is that the first action of the tab is to steer the boat in the ‘wrong’ direction, thus briefly accentuating the yaw before the main rudder responds to bring the boat back on course. Also, on a boat with any kind of inboard rudder, the linkages become dauntingly complex.
Still more power
The story continues with Blondie Hasler. In 1964 he was responsible for the development of a new and immensely powerful type of self-steering gear that was to form the basis of all but a few designs that survive today.
And it was an inspired bit of thinking. Recognising the limitations of the trim tab, he looked for a new way of harnessing the hydrodynamic forces locked into the waterflow. He noticed that if you held an oar over the stern with the blade aligned with the flow, there was a little drag but no hydrodynamic force. But if one twisted it slightly to produce even a small angle of attack, a powerful force developed tending to lift the blade towards the surface – a pendulum action, from which the term ‘pendulum servo’ was born.
Surely, he concluded, by taking steering lines to the tiller or wheel this robust action could be used to steer the boat. And he was right, as history testifies, but there were problems to be overcome first. His early experiments produced alarming results. The servo blade would deflect violently, hard over one way then the other, with the boat yawing wildly in response. To be useful, this essentially brutal principle had to be turned into something more well mannered. There needed to be a taming influence that admitted a powerful initial response that would moderate quickly as the action progressed.
Hasler had been there before with his trim tab gears. It’s all too easy to build a mechanical helmsman that knows only two commands – hard-a-port and hard-a-starboard – and now with the prodigious power his newly developed pendulum servo blade bestowed, it did so all the more dramatically. By contrast, a human crew will take a proportionate view of any course corrections. This involves applying just enough helm to counteract the yaw, gradually reducing it as the boat approaches its correct heading. Naturally, a sure touch on the tiller calls for judgement and anticipation – both products of imagination and intelligence, rather more than you could expect from a machine. Yet, he knew that somewhere in the control geometry there had to be a way of introducing a damping effect that would work entirely automatically. He was thinking about mechanical feedback.
There are various ways of gaining this. A simple example is commonly used in trim tab gears, and is shown left. The push rod from the windvane acts on the trim tab’s tiller at a point astern of the main rudder’s axis. The tab’s action on the main rudder causes it to swing towards the push rod, first reducing, then reversing, its angle of attack. In Scanmar Marine’s pendulum servo Monitor gear (I have one myself – see the header photo) the same ‘positive damping’ is achieved through the master gear linkage, where the meshing of the gears progressively rotates the servo blade back towards a zero angle of attack as it swings to either side.
Variations on a theme
So, a typical windvane self-steering gear is made up of four separate elements:
- A vane to sense the apparent wind direction. Vertical axis vanes are still to be found but by far the majority of manufacturers have opted for the more sensitive and powerful inclined axis type.
- Control linkages. These are usually pushrods, sometimes rotating shafts, and can occasionally be cables. Where the geometry allows it, pushrods or shafts are the preferred choice because there’s usually less friction and backlash – the latter being the slop in any system which can rob it of its sensitivity. Although the linkages might appear to be rather insignificant parts of the whole mechanism, it’s usually here that the positive damping occurs, so their importance shouldn’t be underestimated.
- Control surfaces. These are acted upon, via the linkages, by the windvane. On a trim tab gear the control surface is the tab itself, and on a pendulum servo it’s the servo blade.
- Power output. Trim tabs usually work directly on the rudder blades, though some can be mounted remotely. Pendulum servo gears acting on the main rudder almost invariably use control lines led back to the tiller or a drum on a wheel. This can be a problem on centre-cockpit boats where the routeing can be tortuous.
Not all gears make use of every element. The Hydrovane for instance, uses an inclined vane – necessarily quite large, since there's no servo effect – to operate a semi balanced auxiliary rudder mounted on the stern. Since all the linkages are internal, the unit is entirely self-contained and requires no other connections to the primary steering system.
Other makes also make use of auxiliary rudders. For example, the Windpilot Pacific has a pendulum servo blade controlling its own auxiliary rudder and Scanmar’s Auto-helm makes use of a trim tab to do the same job. As with the Hydrovane, the only connections to the boat are the mounting brackets and the adjustments to set the course.
Although undoubtedly convenient, auxiliary rudder gears have one serious drawback. In heavy weather a ‘standard’ trim-tab or pendulum windvane gear will be steering the boat through the large main rudder, and its effect will be proportional to the conditions. An auxiliary rudder gear, on the other hand, has only its own rudder to exert control, and this may not be enough to keep the boat on track on some points of sailing. To militate this to some extent, it’s possible to use the main rudder to, say, counteract any tendency to round up or bear away, but this still leaves the smaller auxiliary in ultimate control.
Give the gear a chance
Get extra keen discounts and beat expensive shipping costs by buying books through this site! The free delivery service applies to most established destinations worldwide.
If I’ve managed to convey my own enthusiasm for these amazing machines I’ll be well satisfied. Inexpensive they are not but, as crew members, they work tirelessly without complaint and consume absolutely nothing in the way of exhaustible commodities.
Boats and gear Windvane Self Steering - how the wind can steer your boat
