There is a school of thought in the IMOCA 60 world that a good autopilot system is worth sacrificing a few sails for in the overall campaign budget – that’s how big the perceived advantage is for some.
It’s hardly surprising. At speeds that are regularly in the high 20s with sustained bursts in excess of 30 knots, there is much to gain from getting it right and a great deal to lose if you don’t.
At these speeds, the rudder not only alters heading but will behave like an aeroplane’s elevator: every twitch risks changing the fore and aft pitch of the boat. This is already balanced precariously on the giant foil to leeward and the canting keel fin to windward.
So given this, it’s little surprise the human pilot is considered to be the weakest link in maintaining a high speed balancing act. As a result, there continues to be a great deal of effort focussed on designing autopilot systems that would do an F18 jet fighter proud.
You might ask yourself what this has to do with grassroots sailing, but the answer is quite bit – thanks in part to this year’s global shutdown. Unless your crew was drawn from a large family living under the same roof, it’s unlikely you were able to sail fully crewed this spring. And even if you were somehow able to pack the weather rail with crew sitting shoulder to shoulder, who else was out there to race?
So this unprecedented period has at least provided an opportunity to look at the latest autopilot technology and how it could help with both short-handed and fully crewed sailing, either racing or cruising.
Understanding how you can get the best out of your autopilot is more than just simply reading the manual, as I discovered when talking to some of those who depend heavily on their pilots. They all made the point that getting the pilot to work properly so that it can be relied on in challenging conditions takes time.
Article continues below…
How new-age sailing autopilot systems are putting computers at the helm
A couple of decades ago I’d have recommended anyone planning an ocean crossing without a big crew to fit wind…
Can augmented reality really give us a vision of the future of sailing?
Google’s mission statement is ‘to organise the world’s information.’ This almost impossibly ambitious declaration of intent underpins the technology giant’s…
It’s not just racers who can benefit, either. Many cruising crews are sailing short-handed and, as events such as the ARC illustrate, the trend of sailing bigger boats with fewer crew is still growing.
So, being able to rely on your autopilot when you have slowed down to change a sail or put in a reef is just as important when cruising as it is when racing flat-out. Plus, steering in a way that reduces the rolling motion by reading a quartering sea more intelligently could make for a significantly more comfortable ride.
This season also saw a change in the Royal Ocean Racing Club’s rules which now allow autopilots to be used on fully crewed boats. Some argue that a human can perform better than any pilot, but there are plenty of solo rock stars that will tell you quite the opposite. They claim that even some of the most common, mass-produced autopilots can now sail better than a human in certain conditions. The key is setting them up correctly and that, it seems, is where so many of us go wrong.
So we’ve talked to some of the experts in the short-handed racing scene about how they set their systems up to get the best out of them. Getting this right could make life more comfortable, and give you a competitive edge.
Getting it set up
Before you look at working on the pilot’s settings, the first essential step is to calibrate the boat’s basic instruments. “When we get aboard to investigate an autopilot system, it is not unusual to find that the only item that has been calibrated is the echo sounder depth offset,” says B&G product director Matt Eeles.
“Calibration is essential, not just as part of the normal commissioning process, but before any of the data on the displays can be relied on. Here, the compass and the boat speed are fundamental to any instrument system. Without doing this, any adjustments to the autopilot will be meaningless.”
The first calibration is for the compass which, thanks to modern software, is the simplest of all and requires turning the boat steadily through a circle (at around 3° per second), until the display shows that the compass has been fully calibrated.
Next up is calibrating the speed which, according to Eeles, is best done along a known distance between two fixed points. “Use your chartplotter to select a couple of marks that are half a mile apart to set up a calibration run between them. Start the calibration function on the instruments and then make three runs at a steady speed,” he says.
“While it is very tempting to try and adjust your speed based on your GPS output, the results are not that reliable and it is surprisingly hard to get right by this method, especially if there is some tide and wind.”
Working for Boris Herrmann’s Vendée Globe campaign, Team Malizia, solo sailor Will Harris is very familiar with the way in which a professional IMOCA 60 team will go about calibration. But his experience on the Figaro circuit means he also understands the real world aboard far smaller boats with more down to earth systems.
“You can save a great deal of time on the water by studying the various autopilot functions,” he says. “I make a point of studying the manuals and making a list of key functions that I laminate and stick next to the controller as a handy guide.
“When you do get down to the boat there are a few key things you can do before you even head out to sea. First, pick a day or time with flat water, minimal tide and wind. Then, before you do anything make sure you know how to switch the autopilot off in a hurry in case you need to.
“Then it’s a case of finding out whether the rudder and the pilot agree on what dead ahead is. This can be achieved by motoring ahead slowly in the river or harbour where it’s easier to make heading assessments with a backdrop.
“Set the pilot to dead ahead and then watch to see if the boat’s heading remains straight. Make small adjustments until you’re happy. From here, the key function to understand is ‘Gain’ and what it does. Later on you’ll use this 80% of the time to adjust the behaviour of the pilot across a variety of conditions to make it steer like you would.
“More advanced functions like ‘Auto trim’ and ‘Counter rudder’ are functions that you will use 20% of the time to refine your performance later on where you are looking to make the pilot better than you at steering.”
Volvo crewmember and solo/shorthanded sailor Henry Bomby is also well versed in adapting best practice for smaller boats and campaigns.
Sailing with fellow VOR sailor, Hannah Diamond, they finished second overall in last year’s double-handed class in the Rolex Fastnet Race.
Bomby’s starting point is also calibration. “If the boat speed calibration is out, then calculations like true wind speed and direction will also be out, which will start a knock-on effect elsewhere.
“So, whatever the boat, the speed sensor needs to be directly on the centreline and the same for the windspeed and direction.”
Bomby has several other tips.“Clearly it is important to avoid shifty conditions and stay out of any current as best as you can. Then make sure you give yourself sufficient space to perform at least 10 minute runs to settle things down.
“Observation is a key part of the process. Unless you’re about to wipe out or you are in danger of a collision, don’t be too quick to take over – you need to know how the pilot is responding in order to understand what the appropriate adjustment needs to be.
“It’s also important to keep a log of the sailplan and conditions. A square topped mainsail will produce more upwash than a conventional pin head mainsail, so your calibration charts will need to be more detailed. They will also need to take into account when you are reefed and when you are sailing under full sail. Here, the square top mainsail has a big effect.”
Once you have got used to the basic controls and how you vary the adjustments depending on the conditions, getting to the next level of functionality is more about understanding how the pilot delivers the instructions you have set.
“The amount of rudder angle and the way in which it is delivered to the autopilot ram is calculated from the sum of three key factors known as; ‘Proportional’, ‘Integral’ and ‘Derivative’ feedback control, better known as PID,” says Matt Eeles. “On board the boat these factors can be adjusted by the helmsman using three functions.”
“Rudder gain is the first of these three factors (Proportional), which changes the amount of rudder angle used to get back onto course and how quickly it is applied.
Turning the gain up means that the autopilot responds rapidly and aggressively to any request. This is the main function to make sure that the autopilot is responding in a suitable way for the conditions.
“It’s not unusual for people to think that high gain is what they want in all cases, but the reality is not as simple. In a car, if you’re going fast you will want a smaller amount of steering wheel input applied gently to make a change of direction than when you’re going slowly. The amount of gain you may want afloat varies in a similar way.”
“The second factor (Integral) is the Auto Trim function, which learns how much weather helm to apply to achieve a steady heading. Changing this setting adjusts the speed with which the autopilot learns to cope with weather helm.
“To set up Auto Trim, set the boat up on a reach and steer by hand until you are happy with the feel on the helm. Switch the autopilot on and wait to see that the heading remains the same. Now introduce some weather helm by, say, over sheeting the mainsail. If the autopilot doesn’t compensate fast enough you need to reduce Auto Trim to allow it to learn faster.
“The third function (Derivative), we describe as Counter rudder, which is the function that applies rudder in the opposite direction to stop it overshooting the required heading.
“To check this function, set up the boat under engine and note the angle at which you have started and where you want the pilot to steer. Then put in some large course changes of say 20-30° at a time. From this you will be able to see whether the pilot is over- or undershooting. If it overshoots, increase Counter rudder and decrease the setting if it undershoots.”
There are some further points that all three experts mentioned. In particular, take plenty of time available to calibrate, understand and set up your system, don’t expect too much too soon, but do expect to get it wrong from time to time. Getting an understanding of how your pilot works and building confidence in its behaviour takes time.
Top tips for setting up your autopilot system
- Calibrate your compass
- Calibrate your log
- Study the autopilot manual and make cheat sheet/crib card to keep in the cockpit as a quick guide to functions
- Check rudder and autopilot agree on what dead ahead is. Reset as required.
- Understand what gain control does – the amount of rudder angle used to get back onto course and how quickly it is applied. Think of it as you would the steering on a car, less input at high speed than when travelling slowly.
- Under sail, allow yourself space for minimum 10 minute runs under autopilot to allow things to settle down.
- Resist the temptation to take over, (unless there’s a risk of collision or wipe out), observe how the autopilot responds
- Keep a log of conditions, sail settings and calibrations.
- Set up Integral function, (‘Auto Trim’ in B&G terms) to get the pilot to respond to other heading influences such as weather helm.
- Set up Derivative function (‘Counter rudder’ in B&G terms) to ensure that rudder counters correctly to avoid overshooting the correct heading.
First published in the September 2020 issue of Yachting World.