How do you calibrate wind speed?

[leftbrainstuff]

Originally Posted by barnakiel

Is it possible that once your instrument knows how hard 6 knots blows it will be capable of calculating, by proportional calculation: 12, 18, 24 knots?

And so (IMHO) you have answered your own question.

Fair winds,
b.

Extrapolation is not calibration. You can only interpolate to calibrate. Calibration also assumes you have a more accurate device to calibrate against.

Wind speed as measured is not linear. An anemometer measures dynamic pressure from the air which has a known density at sea level it converts it into an angular velocity. You can calculate what this angular velocity for your windspeed range of interest is using simple formulae. You can then spin the anemometer with a drill and timing light. This is how i calibrated my old datamarine anemometer that i had to rebuild myself.

Any sensor when calibrated will have both a usable range and an error range. Often we'll calibrate for a mid range reading and then we'll accept the error at the limits. You can also calibrate to the low or high end.

It sounds like you have concerns you're reading low. Is your anemometer shrouded or experiencing turbulence? Is there friction in the rotating elements?

Sent from my SM-N900T using Cruisers Sailing Forum mobile app

[transmitterdan]

If there is undue friction in the device it cannot be calibrated so no point in trying. Friction not inherent in the design is not linear nor is it time invariant.

[Hydra]

Another problem is the deviation of the airflow caused by the sails: it isn't the same on different points of sailing and it is different when the mainsail is fully hoisted and when it is triple-reefed.

When my company needs to calibrate the readings from an anemometer, we build a scaled-down ship model and we go to a large-scale wind tunnel, to obtain detailed airflow measurements and produce a transfer function between the local measurement and the upstream wind. Even this isn't fully accurate, because it neglects the combination of the marine boundary layer with the ship speed.

And filtering the fluctuations resulting from high-frequency boat motions (mainly rolling and pitching) is another issue...

Alain

[Jim Cate]

Originally Posted by MarkSF

Pick a day with a strong wind, and no tide. Motor directly into the wind, and then directly downwind. The difference in measured wind speed should be double the boat speed.

For example, boat speed is 6 kts. Wind speed is 25 kts. You should measure 31 kts upwind, and 19 kts downwind.

This would work if you were sailing in a wind tunnel, but in the real world, windspeed is constantly varying and the technique fails.

Jim

[Journeyman]

Hydra:
I don't care about airflow effected by sails. I am just trying to calibrate-in believable wind speed magnitude with sails down. I am happy with the wind point cal I have. It's speed I am after. Satisfied to cal that at anchor or motoring with sails down.

Sail crazy:
Not trying to be exact either, I would settle for +\- 10%

Larry T:
No help from RM here. The manual just says use the adjustable offset feature to "match the WS". No mention of preferred sources of "known good" wind speed.

Monte, markSF, Jim cate:
The up/down wind motoring subtraction does not tell you anything about the correct magnitude of the wind.

Transmitterdan:
At the end of the day it's all about "undue friction in the device", isn't it? That's why you have a calibration feature in your display. Different mast head assemblies and mast head assemblies as they age and get dirty will need custom cal.

Weinie
I was afraid you were going to show me that. But removing the cups and display, rigging power to the unit and going for a drive. big hassle.

All:
I was just hoping to hear a good technique for getting an accurate reference wind speed to cal with.

Sounds like I need to buy an expensive handheld wind instrument and go to the mast head while someone below adjusts the display from my voice commands. No wonder display manufacturers are a little vague on this step!

Thanks all.

[transmitterdan]

Originally Posted by Journeyman

Transmitterdan:
At the end of the day it's all about "undue friction in the device", isn't it? That's why you have a calibration feature in your display. Different mast head assemblies and mast head assemblies as they age and get dirty will need custom cal.

I don't think this is right. Friction isn't a "linear" phenomenon so it can't be considered in a linear calibration feature. Once the bearing begins to turn the losses due to friction pretty much go away and rotational speed becomes proportional to wind speed. Friction determines the lowest speed that can be measured but not the calibration at higher speeds. Consider there is a lot more force on the cups turning back into the wind than what is caused by friction.

The reason display heads have a calibration feature is because different vanes have different length arms and possibly different number of pulses per revolution. The length of the arm is what mostly determines the rotational speed vs. wind speed. Friction isn't much a factor. Raymarine vanes output 2 pulses per revolution of the cups. I think some other vanes use 1 or 4 pulses per revolution.