, ,

After last year’s excursion into wild sustained wind readings with dubious sensor calibration, we decided to add a Garmin WND10 wind sensor to the kit.

The nice people at Questus sent me a very old drawing of our very old backstay radar/AIS antenna mount, and I was able to build a mounting block out of starboard at home. It took very little persuading with a wonderbar to get everything to work perfectly.


Now our Garmin Radar/Depth/Chartplotter also shows apparent and true wind. Hopefully soon we’ll be able to use the autopilot to steer to apparently wind angle, which is much safer in some conditions.

The calibration on the old sensor, by the way, is spot on. Last November we had several days at anchor with 40 kts sustained, gusting to 60. Far out.

*****Nerd Stuff Follows – Turn Back Now*****

The first rule of automation is that if a software based system worked once, then stopped working, it’s never the software’s fault. You need to look for a broken wire or dropped connection. When our wind indication got flukey, I was certain I had pinched a wire, which would be a nightmare to find. However, two days of watching showed me differently.

We have two data networks connected to the Garmin: a NMEA2000 canbus and a NMEA0183 serial bus. The wind sensor itself puts the wind info on the NMEA2000 bus. The Garmin reads the info and translates it onto the NMEA0183 serial bus for the autopilot to (hopefully, someday) read. The default settings of the Garmin created a situation where the Garmin would read the NMEA2000 sensor at boot-up, then put the information on the 0183 serial bus, and from then on read the data from the serial bus in a trapped loop.

I finally saw the pattern and found the setting that made the Garmin only read wind info from NMEA2000 and ignore the wind info on NMEA0183. Now everything is perfecto, except for the autopilot communication part.

I told you it was nerdy. You should have listened.