Welcome to the Marine Electronics Forums presented by Panbo and SailBits.
Garmin Reactor 40 and aggressive XTE correction
I just read a Panbo discussion from 2011 , about modern APs and the tendency for aggressive turns to recover XTE error
I just installed a Reactor 40 AP in place of a old Cetrek AP . There’s no doubt the Garmin unit has fantastic heading hold keeping
however in route following , the correction of XTE is truly aggressive , with turns of 45 degrees in closing the xte gap within 100 feet ( with the associated overshoot )
adjusting the rudder gains does little or nothing , it’s fundamentally a design failing
its not not helped by the un-configurable deadband of around 30 feet before the AP attempts track position recovery
some other APs offer adjustments like restart XTE etc to iron out the issue but the reactor 40 offers virtually no user adjustment
i must say I’m not impressed , but its way way too aggressive , resulting in abrupt Z turns about every 0.25 Nm ( ie once the xte deadband triggers at around 30 feet )
I’d be interested in others opinion ?
( surely it doesn’t do this in sailing mode ?)
Dave. I do not believe that your Garmin autopilot, or any autopilot, is meant to work that way.
What I wrote about in 2011 was unwanted behavior when Go-To AP mode was re-engaged after dodging lobster traps or other obstructions:
In that situation it's debatable what the AP should do, and "restart XTE" is definitely a remedy if you don't want to get back on the original course line aggressively. My fairly current Simrad AP will make a fairly aggressive turn to get back on the previous course line if I let it, but it tells me if the turn is more than about 5 degrees, awaits my permission and also offers restart XTE.
But you're describing behavior that the Garmin AP is doing completely on its own, not because you stop and started a nav route, and it sounds intolerable. Which is why I think there must be a way to fix it, though I don't know how.
It does seem very weird Ben
Garmin tech suggest tweaking the rudder gain ( which has been set via the Autotune method ) but all I find that doing is reducing the course hold behaviour rather then significantly changing the overall XTE behaviour
there is very little adjustment available on the Reactor 40 , just rudder gain , counter rudder gain and rudder acceleration limits
none of these change the XTE behaviour in any manifest way
there is no specific “ dodge “ functionality on the reactor 40 , but you can use the buttons to force a course change and the AP goes into heading hold , again on re-selecting GPS steering mode , aka Route following , the AP will execute a very abrupt Z turn to regain the track , which is what you wrote about in 2011 No warning are given about abrupt course changes ( the manual does warn of it ) in fact I’ve had the AP do an abrupt 180 to access a track that was “ inverted “ by me unwittingly , can you imagine that in a sailboat !!!
"I’ve had the AP do an abrupt 180"... yike.
Hopefully someone here or elsewhere can help you solve the sudden turning. But in the meantime, maybe you should use Heading Hold to steer to waypoints. That's what I do most of the time because I'm used to it and there are no AP surprises. Using either the BTW (Bearing to Waypoint) or the graphic course vector on my chartplotter screen I steer toward the next mark, enable Heading Hold, and perfect the course as needed with +/- 1 or 10 degree buttons. I also use those buttons for dodging lobster pots so its easy to return to the original course, and I have BTW, COG and HDG data all visible so that I can personally monitor XTE.
It sounds complicated writing it out but for me it's a natural way to use an AP that keeps me involved in the nav and steering process to a comfortable degree. (And you may know all that, Dave, but maybe its valuable to other readers.)