Robert, Poppy was fitted with racing instruments, giving water speed, true or apparent wind speed and I think direction as well as VMG to up wind or down wind with an expanded scale apparent wind direction from directly up wind or down wind. This latter instrument was very useful and if I remember correctly for beating I simply selected the apparent angle as appropriate to the sea state, trimmed the sails to best speed (all luff tell-tales flying) and stuck to that apparent wind angle. This got the respect of the Bermudan boys.

Unfortunately I cannot remember the exact range for apparent wind angles I used as I used them as pips on the scale rather than numbers, but they could be remarkably small. I do remember that pointing above 45° true wind angle did not really increase VMG to windward, but it was often useful to point much higher if you sheeted in further trying to pass a moored boat or buoy and know that the speed would not drop of significantly and stall the keel. The 30° angle does not surprise me as I have a recollection of tacking through 60° one occasion.

Poppy could point much higher than her Bermudan sister, if for no other reason than the headsail sheeting angle on her sister was right out on the toe rail and the split rig could be sheeted almost to the centre line so that the jibs only had their inbuilt sheeting angle, which was about 10° (I think). The split rig can be sailed very tightly sheeted, but not necessarily with a better VMG. The Longbow is very susceptible to sea state as the bow is quite blunt to give a good fore cabin berth, however in flat water (it does occur occasionally) it was fun to out point all the Bermudan boats. We got some very funny looks at such times.

As for downwind speed, on the run we always got boat speed equal to relative wind speed which meant running at half true wind speed. This held up to 6 kts boat speed, but above that LWL length limitations slowed the increase down. Going onto a broad reach could increase the boat speed in winds under 12 kts, but the VMG down wind would only increase a little so we didn't really bother to tack downwind.

I see little point in comparing the boat speeds recorded for the various boats, even with corrections (or fiddle factors) as a slight variation in sea state can make a mockery of any results, and that is why I have always tried to assess performance when sailing beside similar boats of known handicap (even though handicaps are not perfect). I think the only thing that might come out of this exercise would be to show the shape of the windward performance curve and close windness of a rig, but on Poppy I wasn't worried about that as we had the VMG meter to confirm our performance.

I hope this is of some use.

Cheers, Slieve.

Now that I'm manipulating the date-time stamp just want to make sure about format. Poppy files are obvious dd/mm/yyyy (15/09/2018). Weaverbird format could go either way: 06/08/2018.

Thanks

rself

Poppy prelims. Even TWSs. Histograms, etc here. I also have a split junk rig so found these results particularly interesting.

In Slieve's AYRS catalyst article he mentions that Poppy goes at about half the wind speed. These preliminary results supports that. I point to the TWS=6knt polar plot which, most likely, all 7 panels were up. N=1100 plus data points. Many of those point in the 30,40,50 and 150,160,170 and 180 degree bins, i.e. upwind and downwind TWAs. Median boat speed 3 knts or better in all bins in the 6 knt wind.

I'm wondering about the "30 deg" TWA...is that realistic for a westerly longbow? Seems unusually close-winded for the design but I don't know the boat. There are 200 points in that bin having TWAs between 25 and 34.9 degrees. Are all/most due to tacking? It's fairly easy to look at the time series of d(twa)/dt and d(tws)/dt. Obvious tacks should show up as hills or valleys and "steady state" as flats.....maybe.

rself

Hello Robert,

The CV7 wind sensor is outputting eg

$IIMWV,047.0,R,014.9,N,A*32

where 047.0 is wind direction
R is reference
014.9 is wind speed
N is knots
A is 'correct measure'
*32 is checksum

Also $WIXDR wind temperature which we aren't interested in.

The DX900+ log sensor is outputting eg

$VMVBW,4.65,-1.00,A,,,V,,V,,V*64    dual speed
$YXXDR,A,-3.4,D,PTCH,A,-10.6,D,ROLL*6D pitch and roll
$VMNLA,25.6,A*06     leeway
$VMVLW,38.68,N,28.07,N,,,,*5E    distance
$VMVHW,,,,,4.96,N,9.19,K*5D     boat speed

I think that NavMonPC is then adding date and time, because that sentence is $GPZDA which doesn't appear in the text files. We would need another source if we don't use NavMonPC.

I've put a full list of the NMEA 0183 sentences here

Alan wrote:

 I think it confirms my impression that the approach of taking the median of all the data leads to an unrealistically low result.

It occurs to me that a better approach might be to use an idea similar to the concept of "significant wave height", which is the mean of the highest third of the data, the data being the measured wave height of thousands of waves. This gives a figure that corresponds to an experienced observer's assessment of the sea conditions, but effectively smooths out any extremely large waves, and discounts the lowest two thirds of the data. Maybe we could and should use a similar approach to extract a result from the data that an experienced observer would recognise as "reasonable".

The Pareto Principle, or 80/20 rule, might also be applied. That is, it's easy to get 80% of available performance with 20% crew effort/concentration, but getting that last 20% of performance requires a much higher level of skill and concentration. I would guess that most of us sail at the 80% of available performance level (unless we're racing and concentrating 100%). Sea state, too, will have an enormous influence. In flat water, we might get 100% of available performance, but in wind against tide conditions we might struggle to get 50%. 

Once Anthony's approach has discarded data that is not steady state, because the boat is turning, or the wind is changing, then perhaps a line drawn through the 80% points would be quite realistic, as a guide to what performance is readily achievable.