We have been discussing auto pilot requirements with one of our clients and we thought a review of what makes a good pilot system might be of interest to you.
To begin with, the moderate beam and balanced lines of our hull shapes require very small inputs of rudder to correct course. 99% of the time we could get away with substantially less powerful pilots than the notes which follow indicate. However, it is that one percent of the time, when we are running before a strong gale, or reaching across big seas, that we want the best pilot characteristics. We are willing to help the pilot think, but we do not want to hand steer, ever.
The worse the weather, the more we want the pilot to do the work of helming.
OK, lets start off on a nice broad reach, with the wind at 150 degrees to the boat, 16 foot (5m) waves, and 30 knots of wind. Perfect conditions to make a quick passage down wind under sail or power, if you can control the boat.
For the autopilot this starts with compass response speed. Fluxgate compasses are inherently slow, averaging about 1/2 second on for updating the pilot. This gets worse when headed north. 1/2 second is too slow for good control – the pilot is already behind the curve before it knows to generate a course correction signal. The answer is a fluxgate compass fitted with a heading reference accelerometer, which speeds the updating process to .05 or less seconds, and is not subject to northerly heading error.
Next comes the rudder actuator system, typically mechanical for smaller yachts and hydraulic for larger. Maneuvering in port, and for slow speed steering in heavy weather, fast rates of turn are required. Our standard is ten degrees per second, or eight seconds hard over to hard over.
Normally steering can be achieved with small forces on our hulls. With the 78 foot ketch Beowulf, carrying two spinnakers on a broad reach, average power required was less than 1/16th of a horsepower. Yet we carried half HP pumpsets. This created a bullet proof system, spending most of its life loafing, with the power to provide required steering force in horrible weather. Rudder actuators are no place to skimp.
But high rudder turning speeds can instigate oversteering, and force you to slow down or lose control. Factors which affect this include slop in the steering system and or rudder position sensor linkage, and response speed of the rudder actuator. Dynamic braking can be used with hydraulic systems to more accurately control the rudder and allow higher rates of turn, and some systems can vary rudder response speed.
There are three basic controls set either manually or automatically by the pilot control head. First is sea state (also called dead band). The higher this setting the more the course is allowed to wonder before the pilot starts correcting. Surfing downwind on the FPB 83 Wind Horse we will often open the sea state to five or more degrees, reducing rudder action to a minimum (Wind Horse is very well behaved). On the other hand, when maneuvering in tight quarters this will be set to zero.
Next comes Rudder Gain which controls how much the rudder turns in response to angle off course of the boat. We usually have rudder gain set to its minimum off the wind in nice weather, and set to eight or ten in heavy weather.
Counter rudder helps prevent oversteering with some types of designs (we do not need this).
One of the decisions to be made with autopilot choice is do you want easy and fast access to these three controls? Or, are you content to leave them alone and access them through a computer menu?
Although most of our yachts fit back up hand steering systems, these are almost never used. Even docking or maneuvering in tight quarters the power steering function of the pilot is used. It is much, much faster at turning the rudder than anything a human can do. This raises the question of the type of manual steering control to be used. Our own preference is for what is called a “follow up” controller, where the position of the control knob roughly corresponds to the rudder angle.
How this steering control is turned on when you want to assume control is an important consideration. Think of this in the context of steering the boat in dangerous seas, when you have been at the helm for 18 hours straight, with the pilot in auto mode most of the time. When that breaking sea comes at you from right angles can you hit the power steering button without fail?
We have left the control head for last. Beyond adjusting the pilot response with sea state, rudder gain, and counter rudder, the bells and whistle feature are of little interest to us. We are after reliability and fast control of the rudder.
Which brings us to redundancy. We have been using WH Autopilots for the last 25 years, and they have proven to be reliable in the extreme. Still, we fit two totally redundant systems as standard to every yacht we build. There are two isolated hydraulic systems, two pumpsets, two rudder position transmitters, dual compasses, and control heads. This gear is wired so any component of one pilot can be easily plugged into the opposite system.
If you want to see how all of this plays out offshore visit our online videos. There are some good big wave surfing videos aboard the FPB 83 starting here (be sure and watch parts one and two). Another segment, this time a North Atlantic gale is here. There is also some good Southern Ocean surfing footage aboard Beowulf here.
March 14th, 2010 at 8:44 am
Hi Steve
In the case of Wind Horse (and the 64) there are 3 autopilot control heads – one of which at the helm station is a backup. The other 2 control heads positioned on the fly bridge and the helm station use the same pump set and, therefore, cannot be active at the same time since conflicting inputs would either disable or confuse the pilot in a situation where two people inadvertently gave conflicting command inputs from both stations.
I assume there must be some kind of switching arrangement (button or the like) that must be pressed to activate one station or the other? Couldn’t really find anything on the WH site about this.
Regards
Pete
March 14th, 2010 at 9:27 am
Howdy Pete:
The remote control at the flying bridge is linked to the primary pilot system on Wind Horse (as is the Pilot Wheel inside). WH has these designed so only one of the three can control at one time. The flying bridge unit and Pilot Wheel each have a button to push which takes command.
The back up system is not connected to the remotes. Should the back up system be needed, and yu want the remotes on the back up system, the plugs of these would be switched from one junction box to the other.
The FPB 64 has the same approach to the primary system. The secondary unit, however, has its control head on the inside bridge desk, so it is ready to go (the secondary head on Wind Horse is connected, but stored away).
To change over pilot systems the following is done: select second pilot at the breaker panel adjacent to the bridge; make sure the second pump set has power (this breaker is on the aft panel), open the bypass valve on the hydraulic cylinder of primary pilot which you are deactivating, and close the valve on the secondary pilot’s cylinder.
This sounds complicated, but in reality takes very little time.