Garmin Autopilot Performance

[Eric Jan]

Hi all,

Our autopilot, a Garmin Reactor 40 + Raymarine type 2 linear drive, seems to regularly (several times an hour) go off course by up to 20 degrees either way on a beam reach with say >15 knots of wind and >1.5m swell/waves. And it doesn’t seem to adequately respond, the rudder angle isn’t much when this happens. It just goes of course 10 degrees, waits a bit, minor corrections and then when it hits 20 degrees it slowly (decent speed) gets back to on course.

We installed the Garmin Reactor 40 last winter, the Raymarine Type 2 Linear drive is older but seems to work fine.

We are a 1993 Fountaine Pajot Venezia 42, a 11-12 ton catamaran, and we (still) sail conservatively .

Current settings:
- Power set at 15 amps
- Gain 53%
- Counter gain 67%
- Acceleration limiter 62%
- Tried different response settings (incl. high 5)

My goal for this is to have an autopilot that I can trust in force 7 while putting up the spinnaker, like those solo open 40’s [joke – although I would like that autopilot performance]

What settings (or action) do I need to mess with?

[Eric Jan]

In a facebook group someone shared that it might have to do with the placement of the compass. And that I should try sailing on windhold and see if that is better.

Not sure that is the case, but maybe this update helps.

I now tried/monitored it a bit more and on engine and sailing without to much waves it holds the line perfectly.

But where it starts to have a challenge is with swell/wave action. And in particular when they come on a beam reach or from behind. Then it more frequent seems to swing 15 degrees either side.

Any experts/ideas on this and what setting adjustments to try?

[CSYChip]

I have an older Garmin tied to a Raymarine ram and have found that any play in the bolt that attaches the ram to the rudder yoke will cause the autopilot to go off course. For me it usually starts by turning exclusively in one direction in small increments. Then it will either suddenly turn back or completely spin the boat. My theory is that the play causes the rudder position sensor not to match the calculated expected movement and mess with the calculations. Tightening the ram bolt fixes the issue. I won't say that the Garmin always holds a course as well as I would like but it is better than the Raymarine it replaced.

[Eric Jan]

Originally Posted by CSYChip

I have an older Garmin tied to a Raymarine ram and have found that any play in the bolt that attaches the ram to the rudder yoke will cause the autopilot to go off course. For me it usually starts by turning exclusively in one direction in small increments. Then it will either suddenly turn back or completely spin the boat. My theory is that the play causes the rudder position sensor not to match the calculated expected movement and mess with the calculations. Tightening the ram bolt fixes the issue. I won't say that the Garmin always holds a course as well as I would like but it is better than the Raymarine it replaced.

okay great, I will give that a go... I think the metal treat between the rudder and the sensor is a bit bend so maybe I need to give that a go first.
(just thought that even when bend, the distance stays the same, so why not keep it).

[u4ea32]

Pelagic autopilot. Cheap, simple, very good surfing downwind too.