Wind sensor errors -- source?

[Dr. D]

I was racing yesterday and I thought the the TWA/AWA numbers were a bit odd on a long beam reach. At the first mark I tacked and it appeared the wind shifted as I tacked; well, the instruments said so. After a while I noted a wind shift favoring a tack so I did. After the tack, the wind appeared to have shifted while I was tacking, by some 20 to 30 degrees. However, my sense was there was no actual wind shift. A couple more tacks showed that the wind sensor, or the electronics connected to it, were shifting the wind at each tack. As I was sailing solo, I decided to ignore the instruments for the rest of the race.

What failure of the wind sensor or the electronics might cause this error? Could the wind sensor have been mechanically damaged?

[Plumbean]

Perhaps the effect of wind shear.

[Plumbean]

Some discussion of it here:


https://www.sailingworld.com/how-to/...parent%20wind.

[Dr. D]

I made an error in my original post. It was the True Wind Direction (TWD) that was showing these large swings.

I understand your point about wind shear, though I have not seen this behavior before.

[SailingHarmonie]

Wind shear is not likely to show enough variation that you would actually be able to see/feel the difference, although it is real. Maybe if your mast was 30 meters high, but that's not likely on a Beneteau... It's a good example of looking to an esoteric explanation, when something much more mundane is the likely problem.

FAR more likely is an error in one of the input data streams. All of the following data is used to calculate TWD:

STW (speed through the water)
HDG (Heading)
AWA (Apparent wind angle)
AWS (Apparent wind speed)

An error in ANY of those will result in an issue exactly like you report with the TWD. Nine times out of ten if the funky data is something new the error is in boat speed, especially if you have a paddlewheel speed sensor. One little bit of biology growing on the hull in front of the paddlewheel can upset the calibration.

Other places this can come from: Somebody putting a magnetic wrench (or the whole damn tool bag) near the heading sensor.

Bird roosting damage to the wind sensor.

If you use SOG as a replacement for STW (which real racers don't do unless they HAVE TO) then a current that is a significant fraction of boat speed will cause the same errors.

Of course not all of these will apply to every boat, but it will give you places to seek out issues. Many of the values we are most interested in as sailors have multiple inputs that are easily knocked out of calibration. It's easy to forget that they are all interrelated and if one input is wrong, the whole data train goes off the rails.

[Dr. D]

SH: Next week I will be back aboard and I will investigate each of those inputs for validity. Nice write up.

[Dr. D]

I believe I found the reason for the drifting TWD: A fouled paddlewheel on the triducer (DST810). Here is why:

I motored out of the marina. At first, the STW was zero. After about 10 minutes the STW was showing about the same as the SOG. The main difference as I left the channel was due to current. Once clear of that effect (the current down to 0.3 kts) I began making tacks while still motoring. The TWD plot varied some, as appropriate for the wind conditions, and no large swings with the tacks.

The wind was light, 3 to 6 kts. I put up the sails and shut off the motor. Before I made the first tack the STW went to zero; probably not enough water flow to overcome the resistance. With these winds, the boat was moving at about 2 kts. So I selected the instruments to use SOG for STW to calculate TWD. Now when I tacked, the TWD would swing with my heading. From a port tack, tack to a starboard tack, and the TWD would shift significantly to port. And the other way around.

I also noticed that the instruments were saying the current was now nearly 2 kts, boat speed. What I'm thinking is the AC42 autopilot (which calculates current, TWD, and other terms) uses STW to calculate current. If STW is zero, then the boat is drifting with the current. That then throws off the calculation of TWD.

If you look at the attached image, before the 10 hack, the boat was on a port tack. I tacked onto starboard and the wind shifted the same way. After a few minutes I tacked onto port and again the wind shifted the same way.

My problems with my paddlewheel are due to too much anti-fouling. I should have only applied it to the flat section and the sides. No more on the paddlewheel.