Anonymous wrote:

Robert.

I took the liberty and downloaded the polar diagram of the Contessa 32. This is what I call a true polar diagram. The sailing angle of the boat is no doubt related to the true wind direction. This is how textbooks present them, and also yachting magazines.

I have picked the polar diagram with the wind speed at 8kts as an example and have added two horizontal helping lines to indicate the best VMG upwind and downwind.

My point is that it appears to me that the ‘polar diagrams’ for Amiina and Weaverbird have been drawn using apparent instead of true wind angles. To make these diagrams compatible with true polardiagrams, like that of the Contessa, I am afraid they have to go through another smart algorithm (..which it is too late for me to suggest… a serious exercise in trigonometry is called for...).

I hope this makes sense.

Arne

 

(PS:That polar diagram, below also shows how useless the Contessa’s Bermuda rig is downwind, without a spinnaker  -  or a JR...)

 

You need to be very careful about what you are comparing. This Contessa 32 polar looks to me as though it has been generated by a computer program, and not from measured performance. As such it will represent the best theoretical speed the boat can do.

Robert's analysis takes the median performance value from the full range of data we have recorded. This includes lots of low speed data when the boat was not being sailed well, and so it is inevitably much lower than the maximum potential performance, so his polars cannot sensibly be compared with the Contessa 32 polar.

The polars that Anthony & I produce are based on the maximum recorded performance values, with various corrections and filters to eliminate misleading data, as discussed in the articles I wrote last year. Therefore they are more comparable to the Contessa 32 polar, but still not truly comparable. 

If you want to compare our data with a Bermudan rigged boat, then you should use the Calisto data, recorded using the same instruments, and subjected to the same corrections and filters as our junk rigged testing.

When Anthony has finished running all the boats tested this year and last year through the new smoothing software, I will be placing the results on the website so everyone can see them.

The problem here is that. as we discovered last year, the wind sensor mounted on the transom is affected by downwash from the sail when sailing close hauled.

Last year we developed the idea of correcting the TRUE wind values between 0 and 70 degrees so they are reduced by 10 degrees at 35 degrees off the wind, and the correction tapers off to zero at 0 degrees and at 70 degrees.

This is a completely arbitrary correction to take care of the perceived affect of downwash on the results, and just adjusts the clearly incorrect upwind performance values to something more reasonable.

David Tyler & I discussed the possibility of mounting the wind sensor in the bow just for sailing upwind, but we were unable to come up with a satisfactory scheme in time for this year's testing.

So meanwhile we continue to apply the same arbitrary correction to all the boats so it is at least consistent.

Anthony Cook has been developing smoothing software so we can get sensible looking polars from the jagged recorded data, and has today just sent me a smoothed version of Amiina's polar, which also allows you to see the smoothed and unsmoothed curves next to each other. These polars contain all the corrections which we discussed and published in the magazine last year.

This is an Excel file. Look through the various tabs to see the comparison graphs.

Anonymous wrote:

These polars are based on true wind speed and true wind direction values. In my diagrams the wind is coming out of the west instead of the north and the boats are on stbd tack. The frame of reference is rotated to the right otherwise everything else is the same.

The green dots are the magnitudes of the median vmg's calculated from the actual field data. No drawing of lines is necessary. If you wish you can drop vertical lines to the axes from the red squares but they won't necessarily line up with the green dots 'cause the dots come from the raw data. It's a given that the direction of vmg's are straight upwind or downwind so no need to show direction. The relevant information that I hope is transmitted is boat speed and direction relative to the wind direction (out of the west), the associated magnitude of the related vmg's, and (what you'll never see) the variability of actual field measurements. Less confusing? 

Robert,

in that case, both Amiina and Weaverbird must be very close-winded, indeed. It appears to me that their VMG keeps improving  until they are pointing 30 degrees from the true wind. I struggle with believing that.

Arne

Arne--Yes east.

This is what I see in the plot below.

vmg @ 30 deg course is around 2.1 knots in the straight upwind direction. Just read the magnitude off the radial scale....we already know vmg's are straight up or down wind.

vmg @ 40 deg is higher about 2.3 knts because stw increased but not offset by 10 deg angle change.

vmg @ 50 deg came down a bit to 2.25 knts so increase in stw is offset by 10 deg course change.

vmg @ 60 deg decreases to 2 knts even though stw increase because the course change offsets the stw increase even more.

same for 70 and 80 deg. even though stw increases the length of the vector on horizontal x-axis gets smaller and smaller and goes to zero at 90 degrees because on a beam reach (i.e. north) the boat is travelling perpendicular to the upwind course.

So I see vmgs initially increasing then decreasing to zero as the course of the boat rotates from eastterly (30 deg off the wind stbd tack) to north on a beam reach.

PS-

See text file below. I'd need to understand then manipulate this code in order to change the default orientation. Any volunteers?

Anonymous wrote:

My point is that it appears to me that the ‘polar diagrams’ for Amiina and Weaverbird have been drawn using apparent instead of true wind angles. To make these diagrams compatible with true polardiagrams, like that of the Contessa, I am afraid they have to go through another smart algorithm (..which it is too late for me to suggest… a serious exercise in trigonometry is called for...).

Hi Arne -- See # 6928947 for a furthur explanation of the green dots and red squares.

These polars are based on true wind speed and true wind direction values. In my diagrams the wind is coming out of the west instead of the north and the boats are on stbd tack. The frame of reference is rotated to the right otherwise everything else is the same.

The green dots are the magnitudes of the median vmg's calculated from the actual field data. No drawing of lines is necessary. If you wish you can drop vertical lines to the axes from the red squares but they won't necessarily line up with the green dots 'cause the dots come from the raw data. It's a given that the direction of vmg's are straight upwind or downwind so no need to show direction. The relevant information that I hope is transmitted is boat speed and direction relative to the wind direction (out of the west), the associated magnitude of the related vmg's, and (what you'll never see) the variability of actual field measurements. Less confusing? 

Anonymous wrote:

Just for the record:
1: Are the 'polar diagrams' again being shown using  pointing angles to the apparent wind instead of to the true wind? 

2: Can anyone more qualified than me draw a conclusion from these test sails?

Arne

My polars are all true wind speed (TWS) knts.

I'm working on my version of summary plot(s) which will be similar to Reply # 6936247 on 6922879.

I've put up a prelimnary summary velocity made good (VMG) plot for all JRs including this years Weaverbird Wing and Amiina. Also Alan's tested boat spec spreadsheet showing my computation of Amiina's scalar which, in theory, normalizes differences between the hulls....an attempt to do apples-to-apples comparisons. Hopefully Alan will check my work.

I got a high scalar for Amiina of 1.48 which makes sense relative to the other designs because it has shortest waterline, lowest SA/D of 14, low B/D of 32% but, unexpectedly, VMGs overlap with larger boats. Assuming all Amiina's hull values are the same as a stock Splinter (I used 175 ft^2 sail area for Amiina from Ed Hooper's JRA page) Amiina's VMG numbers should be closer to Miranda's. They are not.

If anyone has access to polar plots for the Splinter 21 design that'd be useful for comparison with the Amiina results.

For comparison you might want to make comparisons to this polar for a Contessa 32 which is a fairly well known class design in Britain anyway....keeping in mind that most, if not all polars you find in mags and online are theoretical output from VPP programs. Actually hard to find polars for small sailboats even designs that have been around for ages like the folkboat.

robert self