Sail GP: Boat Speed vs True Wind Speed

[Doug Lord]

Sail GP: F50s Boat Speed vs True Wind Speed Data | Catamaran Racing, News & Design https://www.catsailingnews.com/2019/02/sail-gp-f50s-boat-speed-vs-true-wind.html

 

[John Perry]

Interesting plot - thanks Doug

A few comments:

I find the way the data is presented is confusing. The graph is labeled BS vs TWS but since the text says that speeds in the 30 knots region were attained the graph must be the other way round - i.e. it is TWS vs BS - not what you would expect - the convention is to use the horizontal axis for the independent variable which would be TWS in this case.

At any TWS the plot shows a big range of boat speed. That is not surprising since boat speed is not a funtion of TWS alone - it obviously also depends on course and on whether the helmsman was actually intending to go fast at the particular moment the data was recorded. So at a TWS of 9 knots, BS could be anything from zero to 33 knots, 33 knots actually being a pretty amazing performance for that TWS. Really, since there is no course information provided by this graph, it is only that 33 knots BS point that is of interest for that particular TWS. Its not really of much interest to know that at that same wind speed the boat often sailed slower than 33 knots, or could even be more or less stationary. So we could consider true wind speeds from zero to 9 knots and plot a new graph of maximum BS vs TWS. However, to do that we would probably need to ignore a few strange data points - for example there is one point that records about 24 knots BS in something like 0.7 knots TWS - that just dosent seem realistic and I wonder what went wrong for that point and a few others that dont look right - presumably some kind of instrument error. Anyway, plotting maximum BS vs TWS for TWS between 2 knots and 9 knots, whilst ignoring the 'outlier points', would give a more or less straight line, at least from 2 knots TWS upwards, its not very clear what happens below 2 knots TWS. A little surprising that this is more or less a straight line, the text indicates that the hulls lift off the water at about 15 knots boat speed but the data does not really indicate anything special happening at that particular boat speed. Maybe the boat is simply moving progressively from a 'foil assist' mode to full foiling so you dont get any obvious speed boost after take-off?

The other point I note from this graph is that there is a fairly sudden cut off in the number of data points recorded for TWS above about 10 knots and data points that were obtained in wind speeds above 10 knots generally show relatively low BS/TWS ratios. I suppose this does not necessarily mean the boat could not perform well at TWS > 10knots, could be that the helmsman just chose to slow the boat down at higher wind speeds.

Anyway, if I take this graph at face value, 33 knots BS at 9 knots TWS is amazing, if it can be believed - a BS/TWS ratio of about 3.7.

 

[John Perry]

A follow on from my post above. When I see graphs of measured sailing boat performance I do wonder about the reliability/accuracy of true wind speed and direction data. Assuming that true wind speed and direction are measured by instruments on board the sailing boat for which the data is plotted, these must be figures calculated from measurements of boat speed, apparent wind speed and apparent wind direction. Boat speed is most likely by gps, that's fine because the likey errors are well known and are minimal compared with potential errors in pretty well any other measurment taken on board a sailing boat. I think there is a particularly big question mark over apparent wind speed and direction measurments. Consider the following:

1) the apparent wind speed and apparent wind direction in proximity to a sailing boat are different to the 'free stream' values. You only have to look at wool tufts or similar placed in a boats rigging to see that. Has anyone actually studied how values recorded by a masthead anemometer compare with values measured at the same height but, say, a few boat lengths distance? I would be interested to see reference to any such study.

2) TWS and TWS direction both vary quite rapidly with both time and location and in an unpredictable manner,

3) Most wind instruments on sailing boats do not even attempt to compensate for boat heel - not so important for multihulls but could be important for monohulls.

4) Industrial instruments for measuring air flow can often be in error by at least 10 per cent. Within 5% is probably as good as can be expected. Most sailing boat wind instruments are probably no better and could well be worse, although developments in instruments without moving parts may be an improvement.

 

[Doug Lord]

Interesting comments- thanks, John!

 

[Doug Lord]

Kyle talks wing trim:

 

[fastsailing]

A follow on from my post above. When I see graphs of measured sailing boat performance I do wonder about the reliability/accuracy of true wind speed and direction data. Assuming that true wind speed and direction are measured by instruments on board the sailing boat for which the data is plotted, these must be figures calculated from measurements of boat speed, apparent wind speed and apparent wind direction. Boat speed is most likely by gps, that's fine because the likey errors are well known and are minimal compared with potential errors in pretty well any other measurment taken on board a sailing boat. I think there is a particularly big question mark over apparent wind speed and direction measurments. Consider the following:

1) the apparent wind speed and apparent wind direction in proximity to a sailing boat are different to the 'free stream' values. You only have to look at wool tufts or similar placed in a boats rigging to see that. Has anyone actually studied how values recorded by a masthead anemometer compare with values measured at the same height but, say, a few boat lengths distance? I would be interested to see reference to any such study.

2) TWS and TWS direction both vary quite rapidly with both time and location and in an unpredictable manner,

3) Most wind instruments on sailing boats do not even attempt to compensate for boat heel - not so important for multihulls but could be important for monohulls.

4) Industrial instruments for measuring air flow can often be in error by at least 10 per cent. Within 5% is probably as good as can be expected. Most sailing boat wind instruments are probably no better and could well be worse, although developments in instruments without moving parts may be an improvement.

I think you might have missed the cause having the greatest effect on calculation of true wind speed and direction. Leeway angle of a foiling boat going real fast and then flying a bit too high with suddenly increasing leeway. The instrumentation will see a rapid and large reduction on apparent wind angle (relative to the centerline of boat not to the direction of boat velocity vector as it should use) while the boat speed is almost constant and there is only a very small reduction of apparent wind speed. The calculation wrongly interprets this as substantially reducing true wind speed together with maintaining unbelievably high boat speed.
In the real world true wind speed remains the same, with substantially greater value than the false result of the calculation.
To get a correct result it is very important to receive true direction of apparent wind vector, not calculate it from apparent wind angle relative to boat centerline (which is calculated based on measured mast rotation and awa relative to mast longitudinal axis) without taking into account how much leeway the boat really has.

 

[bjn]

I think you might have missed the cause having the greatest effect on calculation of true wind speed and direction. Leeway angle of a foiling boat going real fast and then flying a bit too high with suddenly increasing leeway. The instrumentation will see a rapid and large reduction on apparent wind angle (relative to the centerline of boat not to the direction of boat velocity vector as it should use) while the boat speed is almost constant and there is only a very small reduction of apparent wind speed. The calculation wrongly interprets this as substantially reducing true wind speed together with maintaining unbelievably high boat speed.
In the real world true wind speed remains the same, with substantially greater value than the false result of the calculation.
To get a correct result it is very important to receive true direction of apparent wind vector, not calculate it from apparent wind angle relative to boat centerline (which is calculated based on measured mast rotation and awa relative to mast longitudinal axis) without taking into account how much leeway the boat really has.

Good points. And from the pictures it looks like the anomometer is located at the end of the bowsprit. So the apparent wind angle is probably much more forward, or much lower angle than higher up in the sail. Which makes the calculated true wind speed more sensitive to effects like leeway as you say, or steering. A fast change in direction will probably record a much higher or lower TWS than in a straight line, due to the location of the sensor.