I've been away for a few days and also had to get my boat hauled out and transport her home. She was buried under a huge pile of snow for 2 weeks and I was unable to get through the main hatch to get inside the cabin. When it was warm, before the snow:
Hope you had a nice Christmas and all the best for the new year.
It's a pity I'd not already purchased the fluxgate because I could of had a play. Theory and practice don't normally agree! In fact, many years ago a tutor said to me "When Theory and practice agree . . . worry"
It would be interesting to actually see the variation using the Fluxgate hooked up to the Arduino and an LCD display! I then thought what about using a cheap PCB GPS module. There should be little if no variation but the unit could still be gimbaled in order to keep all the current satellites in sight of the aerial. Then I had yet another thought . . . "AVERAGING"
Suppose my intended course is 260º. The fluxgate is varying between 255º and 265º. The Arduino could perhaps take readings over say for a 10 second period, take an average and then respond to the Servo? As it's wind and wave that are affecting the actual course, I doubt if we'd end up with a symmetrical ± figure. Perhaps more like + 7 and − 3 or whatever. To compensate for this, there could be two additional buttons on the unit that are + and - 1º to the stored number (heading in º) for every push of the button. The entry would then most likely remain true for the period of that Tack until I change the boat to another tack (zigzagging across the wind).
I have to do some urgent work on the boat. All 5 windows need removing as they leak slightly. They are all surface mounted and bedded onto mastic. Over the years it has gone hard and cracked. All need re-bedding on new mastic/sealant. When I've got this out of the way, I'll purchase a Fluxgate Compass Module and take it from there.
Bye for now.