[g0rsq]

Hi,

I can't get the full datasheet to download, but the way that I read the (Digikey) specification is that it has "endstops" at 10 degrees and 360 degrees, so would be useless in a wind vane.  I believe the "Continuous" refers to the "infinite" (analogue) resolution between those endstops, i.e. dependent on the resolution of the (external) ADC, or ultimately the noise level of any of the electronic components in the system.   Not in the way the term is used in model makers' "continuous rotation servos".

Of course it is possible to make (or buy) a sensor which measures the (signed) X and Y (i.e. Easting and Northing) magnetic field strengths from a strong rotating "bar" magnet and calculates a direction (see the "ATAN2" function), but I suspect that calibration (in the influence of the Earth's magnetic field) may be an issue.  There are lots of different approaches to measuring the wind direction, which rather suggests that there is no easy or perfect solution.  

Cheers,  Alan.

I have read the spec sheet, and it looks to be perfectly usable!

AMS22B5A1BHASL136N

AM       Analog Magnetic
S         Single turn
22       22mm
B         1/4" shaft
5          5 Volt
A1       12bit resolution
B         0.5% linearity
H         200uS processing speed
A         1-99% of supply voltages as output voltage
S         Slotted Shaft
L         Axial
1         Output volts increase in CW direction of rotation
36       electrical angle is 360 deg
N         Continuous rotation, no hard stop


4096 points digital resolution gives 11.3 voltage points for each degree of rotation

I might give this a look, as I find the fineoffset a bit of a compromise.

Peter