A boat’s autopilot is its most expensive electronic component. So it is important to know it well and know how to use it to get the best out of it. However, autopilots have not always been electric, their evolution has required years of refinement. At the dawn of ocean sailing there was nothing but windvane pilots. Being mechanical and not dependent on electricity, they have been the heart of offshore sailing for decades.
However, the windvane pilot has limitations that do not make it ideal in all circumstances. The world of races has therefore given impetus to the development of electric autopilots. At first they were very simple, but a modern autopilots are really very sophisticated. On modern racing boats, windvane pilots are no longer seen, however they still have a large market in the cruisers’ world. In this article we try to understand its pros and cons and evolutions over the decades.
Mechanical windvane pilot and electric autopilot
Before electronics invaded our boats, there were only mechanical autopilots. A mechanical autopilot is called a windvane pilot. In English they are called windvanes or windpilots. Their story is very fascinating, before a commercial version was produced they were all self-built. At the first edition of the OSTAR in 1960 all competitors had one of their own engineering.
The windvane pilot
When Francis Chichester completed his first sailing circumnavigation in 1967 he became a hero. Sailing with a sextant and a windvane on the stern of Gipsy Moth IV he became a legend. The following year the Golden Globe started, the first non-stop round-the-world race. Robin Knox-Johnston became the first man to complete a non-stop sailing circumnavigation. Bernard Moitessier was also at the start and his photos with the sextant in hand made history. His boat was also equipped with a self-built system.
The principle of operation of the windvane pilot
The operation of a windvane pilot is more complex than you think. These systems too have evolved over time trying various solutions. There are effectively two methods, one that uses a servo-pendulum blade (e.g. Monitor) and one that drives a secondary boat rudder (e.g. Hydrovane). The sail, or blade, of the windvane rudder certainly cannot have the strength to steer a large boat of many tons and in such cases those pilots that use a secondary rudder need a twin pilot installation. On a servo-pendulum system the small oscillations of the blade are trasformed into a force sufficient to steer the boat but work best with tillers rather than wheels and their installation is not always possible.
To do this, the airblade is connected to a blade immersed in the water that looks like a rudder. However, it must be immediately clarified that this blade does not act as a rudder on servo-pendulum windpilots, it is so only on winvanes like the Hydrovane. On servo-pendulum systems, the blade is mounted on a vertical tube with a fulcrum point and can swing from left to right. On a Hydrovane or auxiliary rudder system the rudder steers the boat and does not oscillate. On servo-pendulum systems what makes the blade swing is the airblade, causing it to rotate slightly, but this does not alter the course of the boat, as the pendulum is connected then to the boats rudder. On auxiliary rudder the same rotation turns the auxiliary rudder to steer the boat directly being independent of the boat’s rudder. The airblade is adjusted to a certain angle to the wind and remains vertical in the absence of other forces. When the wind instead of flowing along the axis of the blade hits it sideways, it knocks it down on one side. The airblade in the air has a counterweight, so even a little air is enough for this to happen.
The blade oscillating causes a rotation of the axis of the immersed blade. This is either to generate a force that will be used to move the boat’s rudder on servo-pendulum systems or to steer the boat directly on auxiliary rudder systems. On servo-pendulum systems the rotation causes the blad to swing left or right by the flow of water that hits it. On auxiliary rudder systems the blade acts as rudder and steers directly the boat. Water being much denser than air has a much higher strength, this one principle that has led to the development of servo-pendulum systems, which transform a small force into a large force by the difference of density of air vs water. On auxiliary rudder systems a very well balanced rudder blade mades it possible to steer even a large servo-rudder with little force, it is all down to the use of a properly balanced neutral rudder blade. On a servo-pendulum system the two lines tied to the tube which hold the servo blade are then led back to the tiller. On auxiliary rudders systems such as Hydrovane there are no further lines or complications. On a servo-pendulum system, the two control lines are usually tied to a chain which has coupling point on the tiller bar. The chain allows for fine adjustment if you want to keep the bar slightly off-center. On an auxiliary rudder system the main rudder is not used to steer the boat, rather it is locked in place and can be used as a trim-tab to make the boat as neutral as possible to the windvane auxiliary rudder so that it will need little forces to steer the boat.
The pros of the windvane pilots
The flow of water that hits the blade immersed in water is capable of generating an enormous force – this is very important on servo-pendulum systems that have to steer with the original boat rudder. Less important for the auxiliary rudder systems as the boat’s rudder is not used to steer the boat. With servo-pendulum systems the force of the water is transmitted to the lines led back to the cockpit thus capable of moving the tiller. The second strong point lies in the fact that these systems have no electrical component. As long as it does not break down, we have to worry only about accidents such hitting semi-submerged objects or breaking the airblade.
The blade in the water is usually connected to the main pipe with an intentionally weaker section of pipe. In the event of a collision with an object the blade will bend the sacrificial tube section. The blade itself, tied to the boat, will not be lost, and it will be possible to replace the sacrificial tube. As for the airblade, this too can be damaged in very strong winds. Being very light just bring spares and the problem is solved.
Apparent wind and beating with a windvane
The enormous strength of the windvane pilot is to be able to steer even boats of important tonnage. The forces involved are not indifferent and it is necessary to buy a system of the appropriate size for your boat. However, it is truly impressive to observe the ability of one of these winvanes to steer a boat even in a storm.
By definition, the blade in the air responds only to the air hitting it. The windpilot can therefore lead a boat only relative to apparent wind. For traditional displacement sailboats this does not present major problems at any speed. However, it is important that the boat is well balanced and you can say that your windvane will teach you how to balance and handle your boat better. The strong point of the windpilot however remains when beating or broad beating, where it really gives its best. With a strong apparent that controls the blade precisely, the boat responds and sails very well.
Weaknesses of windpilots
The large forces involved can cause the windvane to break. Damage to the blades in the air or water are easily remedied. Any damage to the mechanical parts are difficult to solve. The structure itself is susceptible to damage in particularly harsh conditions. Some models have an aluminum body and if this breaks it will not even be possible to weld it later. For those in steel, even though you might not be able to fix them at sea, it is always possible to repair them later.
The other problem is the possible absence or lightness of the wind. In a flat calm the windvane cannot work, by definition and it can struggle in light variable airs. In light airs we are forced to steer by hand, until there is enough air to re-engage the windpilot.
Sailing downwind with the windvane pilots
The ability to steer with precision of windvane systems is somewhat diminished when sailing from broad reach to downwind, when the apparent wind is less. Also because the boat is subject to accelerations, especially due to the rising and falling from the waves. As this happens the apparent wind changes. When we accelerate the apparent goes forward, when we slow down it goes aft. In other words, apparent wind swings fore and aft when sailing downwind with respect to its average angle.
This oscillation causes the boat to follow an “S” shaped course that is all the more marked the more important the accelerations. For this reason the windvane pilots are still a valid solution on displacement boats but cannot be used on modern racing and planing boats. These have accelerations that are too sudden, making it difficult for the windpilot to work accurately. The lightness of the boat also implies a lack of inertia, and speed inevitably undergoes continuous and important variations.
When we add these variations in the speed of the boat to those of the wind, everything becomes more complicated. With the apparent swinging continuously relative to the boat, the windvane pilot is unable to steer accurately. Especially on a planing racing boat sailing downwind. The only way to manage with a windpilot on a planing hull is by greatly reducing mainsail and keeping canvas forward and accepting the slightly drunken couse sailed by the windvane. On the other hand, on a non planingn hull, a sailing only with a poled out jib can also be steered by windpilot in large seas.
Leading manufacturers of windvane pilots or windpilots
- Aries (one of the most popular, but aluminum body and steel parts sometimes give problems)
- Hydrovane (in production since 1968 – Global Solo Challenge Event Partner)
- Monitor
- Windpilot (the rudder is called Pacific, and in its Pacific Light version it is also suitable for small boats)
Electric autopilots
Electric autopilots have undergone a slow evolution over time. Modern ones are very complex electronic objects that have nothing to do with the past. The first electric autopilots introduced possibility to steer to a compass heading. The first autopilots, far from being control units, were a bit like stupid mules. They just corrected heading left or right when the compass heading deviated from the set heading.
At the beginning, however, even in their simplicity they provided a secondary system to steer the boat. Above all, they solved the age-old problem of steering in a flat calms. Electric autopilots has no problem keeping the boat on a certain heading. This may leave us to trim the sails in irregular shifty winds.
The anemometer
On board boats the first anemometers (wind sensors) appeared, which detect intensity and direction of the apparent wind. It was only a matter of time before wind sensor (the anemometer) were interfaced with the automatic pilots. This way the autopilot could steer based on the compass heading or based on the apparent wind too. This was a small revolution because on paper now a sailboat electric autopilot did everything the windvane pilot did. Plus he knew how to steer the boat on a straight line in light airs.
The South Atlantic without wind sensor
During my participation in the 2011/2012 Global Ocean Race in the middle of the first leg we found ourselves without a working wind sensor. We had two masthead wind sensors but the first was damaged by a squall in the middle of the doldrums. When we were still 3,000 miles from Cape Town, the second wind sensor tore itself off the masthead. An inspection revealed that the steel support had failed. Over time we discovered that it was a very common problem but the supplier refused to accept any responsibility. A perfect example of bad after sales.
The wind sensor lay thousands of meters underwater on the bottom of the South Atlantic. The distributor asked me to ship the defective part in order to fill out the return form. The person knew where I was and that I had nothing to return – but the world is full of clever people. Not sure that’s how you ensure you have a good reputation as a supplier.
This left us with a huge problem, 3000 miles and only the compass mode of the electric autopilot. During the day my co-skipper steered with an incredible dedication. From dawn to dusk without complaining he managed to remain concentrated and absorbed in his eternal steering. For him it was second nature, a job he tackled with the systematicity of a truck driver who grinds miles after miles anytime and anywhere.
The problem of the night
Unfortunately, Paul was unable to steer with the same efficiency at night. The lack of light at night did not make him see the Atlantic waves well. I tried it too but with rough seas and pushing hard under spinnaker we ended up wiping out lots of times. Moreover, the night was long and cold in the roaring forties and I was forced to devise a way out of that sticky situation. Sitting at the chart table with the remote control in my hand, I could feel the movement of the boat with my body.
Every time the boat “raised its stern” I went 10 deegrees lower When she sat flat I went 10 degrees higher, I stayed like that all night, remote control in hand. The situation was so absurd that we coined the term “Tamagotchi Sailing“. With that remote in my hand, oval shaped like a Tamagotchi, I seemed obsessed with that old Japanese game. Actually what I did was use the compass mode but adapt to the slight fluctuations of the real wind.
Electric Autopilots: the introduction of true wind mode
With the evolution of automatic pilots, in addition to the wind sensors, boat speed sensor was also interfaced. The displays in the cockpits have been showing real wind information for some time. With ever lighter and more planing racing boats, the ability to steer in real wind mode became a requirement. A displacement boat has its own inertia and maximum hull speed. For this reason, its accelerations are neither very sudden nor far from the average speed.
On a racing boat, surfing speed can be double the average speed, and decelerations are also sudden. A racing boat is much more nervous in this regard, due to its light weight. Automatic pilots that introduced the possibility of steering to the true wind angle solved many problems. Racing boats could for the first time sail safely and at high speed even downwind. This is because doing a vector calculation the autopilot brain knew both apparent and true wind angles.
The systems evolved continuously and the automatic pilots became real electronic control units. With ever more sophisticated intelligence, accelerometers were added to provide additional information to the autopilot. A modern autopilot, if well set up, manages to steer perfectly in all conditions. Anyone who thinks it’s not the case often just doesn’t know how to balance the boat well. That is, even for electric autopilots the boat must be well balanced.
Electric Autopilots: the main manufacturers
- NKE Marine Electronics (French, the first to spread in the world of racing)
- B&G (American, is progressively taking market share from NKE which no longer predominates as it once did)
- Raymarine (Designed for the cruising market, it has reached a decent level for use among non-professional racers)
- Garmin (Their autopilot is relatively unknown in the racing world)
- Navman (This autopilot is also not very popular among racers)
Installation of an electric autopilot
The installation of an autopilot, given the interconnection with the sensors, also affects the instruments. For this reason, once the decision has been made on the brand, it will not be possible to easily change your mind without reinstalling everything. They are not compatible with each other (save in rare cases), we cannot take data from a system and pass it on to another autopilot.
The only exception is that of the drive, i.e. the arm that controls the rudder. This is no longer in the cockpit but safe below deck and usually connected directly to the rudder shaft. The actuator proposed by NKE for example is made by other despite carrying their brand. The manufacturer Lecomble & Schmidt which produces excellent and reliable hydraulic actuators. However, the hydraulic actuators have quite high electrical consumption. It is possible, and indeed preferable, to install a Linear Drive Type 1 or 2 (depending on the boat) from Raymarine.
These electric actuators are truly indestructible and can cover tens of thousands of miles without maintenance. Small boat actuators operate on 12v reversible current. Therefore you can always install an actuator of one brand with another autopilot. The combination of NKE brain and Raymarine actuator is the one I went around the world with. I had the same configuration on board my Mini 650 Basecamp 438 winner of the Mini Transat in 2005.
Parameter settings of an NKE autopilot
In a future article we will talk about autopilot settings. Here a brief overview of NKE. In fact, I learnt over time how to make the most of the settings, you need to learn your settings for your boat. Here is a reference table that I used in the Offshore Sailing Training Centre I used to run. For many people the table below will be gibberish others will read on with ineterest.
| NKE pilot parameter settings diagram on a Mini 650 Series | ||||
| Point of sail | Pilot mode | Gain | Speed Cofficient | Wind damping |
| Under engine | Compass | 2 | 5 | n.a. |
| Very light wind | Compass | 2 | 5 | 2 |
| Flat sea upwind | Apparent Wind | 2 | 4-6 | 3 |
| Upwimd rough sea | Apparent Wind | 3-4 | 5-7 | 3 |
| Upwind / flat across the sea | True Wind | 3 | 6-8 | 1 |
| Broad beat / Reach with waves | True Wind | 4-5 | 6-10 | 1 |
| Downwind in a flat sea | True Wind | 4 | 5-8 | 1 |
| Downwind with waves | True Wind | 5-6 | 7-14 | 1 |
| Note | ||||
| The counter rudder must always be set to Auto.The speed coefficient must be proportional to the speed of the boat with a formula equal to approximately Speed Coefficient = average speed of the boat in knots. Then add +2/3 with significant waves. | ||||
| Sailing at 6 knots the speed coefficient will be set at 6 with flat sea and at 8/9 with waves, at 10 knots at approximately 10 with relatively flat sea and at 12/13 with waves etc. | ||||
Parameter settings on other boats
On the Mini 650 Prototypes the speed coefficients must be slightly higher. For example, you can start with a boat speed factor multiplied by 1.2. So at six knots with flat sea we will have it at 7/8. This is to take into account the greater reactivity of the lighter proto.
On a Class40 we multiplied boat spead by 1.5. At 10 knots the speed coefficient in relatively calm seas was at 15. The experience and knowledge of your boat will lead you to adapt the table to your needs.
