Boat speed calibration mysteries

[Lodesman]

Originally Posted by LakeSuperior

I didn't read all the responses in detail so I may be repeating.

You need to recognize that the true velocity vector is the vector sum of the forward velocity vector and the slipage or leeway velocity vector plus or minus 90 degrees from forward velocity. This true velocity vector, speed and direction, is what is measured by the GPS using doppler.

Your paddle wheel boat speed sensor is only measuring the forward velocity vector. This paddle wheel sensor does not respond to the leeway speed. For example, if only leeway and no forward speed the paddle wheel would measure zero speed. I don't think the paddle wheel errors because of the heel however the leeway would be correlated to the heel angle.

Leeway speed will vary between zero and some amount depending on your boat, wind speed, and how you are trimmed.

Consequently, you will see a varying speed mismatch depending on the slippage speed along with all of the other calibration, surface current, and GPS data damping errors.

The OP had touched on this on the first page. If SOG vector is sum of STW and Slippage vectors, then it will always be longer than STW (ie. it's the hypotenuse of the vector triangle). OP reports the opposite - STW > SOG.

[LakeSuperior]

Originally Posted by Lodesman

The OP had touched on this on the first page. If SOG vector is sum of STW and Slippage vectors, then it will always be longer than STW (ie. it's the hypotenuse of the vector triangle). OP reports the opposite - STW > SOG.

Agree! Then to the next level, with wind current, waves, and calibration error on the paddle wheel sensor, the result observed by the OP could be accounted for.

IMHO, SOG and COG measured by the GPS is the most accurate data set available to the skipper. The precision of the remainder of the boat sensor measurements is much less, relatively unknown, and suspect.

For example I imagine a speed dependent error on the paddle wheel due to flow acceleration over the hull and turbulence across the paddle wheel pocket.

[mglonnro]

Originally Posted by Lodesman

The OP had touched on this on the first page. If SOG vector is sum of STW and Slippage vectors, then it will always be longer than STW (ie. it's the hypotenuse of the vector triangle). OP reports the opposite - STW > SOG.

Originally Posted by LakeSuperior

Agree! Then to the next level, with wind current, waves, and calibration error on the paddle wheel sensor, the result observed by the OP could be accounted for.

IMHO, SOG and COG measured by the GPS is the most accurate data set available to the skipper. The precision of the remainder of the boat sensor measurements is much less, relatively unknown, and suspect.

For example I imagine a speed dependent error on the paddle wheel due to flow acceleration over the hull and turbulence across the paddle wheel pocket.

The "mystery" data was indeed recorded in (virtually) zero-current and zero-wind conditions, driving in straight lines back and forth. I think that the most likely solution to the mystery part (counter-intuitive difference when accelerating and decelerating) was related to some internal averaging functions in the instruments themselves.

When I redid the test (and re-checked the numbers), I ended up with a linear gain correction equation for the boat speed (as measured), and I think that is as good as it gets for the base correction.

The next problem, however, was how to correct the paddlewheel speed for heeling (and looking around the internet, some people do!). Not as a correction for the missing longitudinal speed (as some of it turns latitudinal due to leeway), but as a correction due to the different conditions surrounding the paddlewheel itself.

Since there's no mathematical formula for this, I guess the actual problem is to
figure out a reasonably doable test scenario to test this?

Maybe, if I could sail both ways on a track and record the (average) heeling as well as the actual leeway, I could somehow separate out the heeling effect once I have adjusted it for the leeway vector?

It all feels very theoretical at the moment, though, and I wonder if reality is too chaotic to make these kinds of calibration runs.

I think the initial gain correction equation was a big step in the right direction, and it might be that the boat speed correction is "good enough" for the moment. (The end game and desire is to produce comparable real-life sailing performance data for our boat.)

[Paul L]

Mg,
Since you seem to enjoy acquiring data, you could do something like this for the heel sampling.

Go back to your still, calm, no current day. Push the boom way out on one side and load it with weight, such as water, or dinghy. Move extra sails, etc to the same side. Maybe fill or empty a water tank. In the end you will have some fixed heel. Do your speed tests as before including 180* runs.
Then do the same on the side and retake data.
This will give you 3 points of heel, x* to port, 0* and x* to starboard. All without a leeway complication.
Evaluate the data and see if it hints at needing more involved heel correction or not.

[mglonnro]

Originally Posted by Paul L

Mg,
Since you seem to enjoy acquiring data, you could do something like this for the heel sampling.

Go back to your still, calm, no current day. Push the boom way out on one side and load it with weight, such as water, or dinghy. Move extra sails, etc to the same side. Maybe fill or empty a water tank. In the end you will have some fixed heel. Do your speed tests as before including 180* runs.
Then do the same on the side and retake data.
This will give you 3 points of heel, x* to port, 0* and x* to starboard. All without a leeway complication.
Evaluate the data and see if it hints at needing more involved heel correction or not.

Lovely idea!

[LakeSuperior]

An aside here.

I think one could measure leeway independently and more accurately by not using a compass.

This approach is contrived but FWIW set up a buoy with a known position. Then given a recorded starting position head to the buoy. So now you have the hull heading as the course between these two points. Record the paddle wheel speed (the hull speed for the hull speed vector) and GPS SOG and COG.

From this data you can decompose the true GPS velocity vector into the components of the hull and leeway vectors.

Other errors would be minimized if the starting point range to the buoy or fixed point is reasonably far away like 1000's of yds.

Step and repeat using varying points of sail, wind speed, sail trim, and heel.