rob, paul(s): thanks for the info. that's the way i thought wind vanes worked. i guess i still don't really see the problem.
keeping a constant AWA is how i manually steer. at least when close hauled or trying to maximize downwind VMG. on a beam reach, you're right, manually i steer a compass heading and trim the sail.
but if you choose to give up precise control of either heading or trim, which is faster on a reach: oscillating about the right course and keeping the sails in trim, or keeping on course exactly and oscillating in and out of trim? i don't know.
a draft energy budget on my fairly spartan dream boat has the compass autopilot consuming 40% of the amp-hours. a windvane would have to have some serious disadvantages to make up for the savings in weight from less batteries, solar panels, fuel for the alternator, etc.
rob's worst case is of course bad, but if your hand is not personally on the tiller, you shouldn't be canvassed such that that could happen anyway.
perhaps the problem with wind vanes is that there is not enough damping in the feedback. or that the error signal changes with boat speed, as suggested by paul. either would make it a control theory problem. would just need to figure out how to mechanically change the proportional, integral, and differential (PID) coefficients to smooth out the oscillations a bit in all conditions.
rick's self-trimming wing sail perhaps makes it all moot. with a compass autopilot keeping the heading constant, and the vane on the wing keeping the sail trim optimized, the only compromise you'd have to make is the extra energy consumption. hmm.
here's an idea: use the compass to adjust the water servo instead of a vane! switchable between the two so you can choose whichever suits the conditions...
ben
> The apparent wind problem is that as the boat speeds up, the wind operated
> auto helm causes it to bear away to keep the same apparent wind angle. The
> fear is that if you go too fast and slow down suddenly (into the back of a
> wave) the apparent goes aft and causes a capsize.
> I have also seen windvanes on multihulls but the ones I have seen would probably
> rarely get over 15 knots. I think the problem is the shift in apparent wind with
> varying wind speeds but also the foil design. I think it is a real challenge to have a foil large enough for slow speed but small enough that it doesn't fly apart at high speed. At high speed (>15 knots?) you can get severe vibrations, cavitation, and over steering since the amount of rudder deflection required to turn the boat is much less at high speed. It is also always a trade-off with windvanes between being light (delicate) but responsive and heavy (strong) and unresponsive. This is big challenge with heavy monohulls, but an even bigger challenge with high speed multi-hulls.
> That's the worst case. More common with wind vane self steering would be the wild course changes as the boat accelerates and the apparent wind comes forward so the autopilot bears away to keep the same apparent wind angle and then slowing down causes another radical course change as the wind goes aft.
> Monohulls don't get the sudden accelerations and decelerations so a wind vane can steer a more steady course.
> When using self steering a steady course is usually desirable and this can be achieved with a wind vane on a relatively slow and heavy cruising mono.
> On a relatively light and fast multihull a wind vane will not be able to steer a steady course so a compass controlled self steering would be a more successful option to steer a steady course, although with the disadvantage of not maintaining max speed as the apparent wind comes forward.
> When I had an autopilot I mainly used it while motoring, a situation where a wind vane would not do the job and of course the engine was supplying the electrical power to drive the autopilot.
> Cruising sailors would probably want to use a self steering system for much of the time when sailing but I think only a slow heavy cat would be able to successfully use a wind vane. A harryproa would, I think, be to light and fast to use a wind vane. I think a simple tillerpilot would be the best. just disconnect it to shunt and reset it when sailing on the new tack. I never had any success with the autotack function anyway.