Sail measurements

Home Design Build Race Links Reports Other Topics


Larry Robinson alerted me to the UK Sailmakers release of a free software download (http://uksailmakers.com/accumeasuredl.asp the last time I looked) that helps you take some basic sail measurements from your digital sail photo. The screen shot shows two shape stripes measured from one of my Carr HTs, set in about 3 knots of wind.

(Please note that I've not been too careful in this screen shot about aligning the spline curve with my draft (shape) stripes. Note also that the splines are placed over the shape stripes in the photo, but the dark curves near the splines are in fact the seams of the sail.)

As shown by the screen shot, the middle stripe shows a maximum camber of 9.8%, positioned at 38.3% of chord, and has a relative twist of "109.2". This twist measurement means nothing by itself. It only is useful by comparison with the twist measurement of the bottom stripe.

The bottom stripe shows a maximum camber of 9.1% at 32.3% of chord, and a twist of "116.2". The difference between the two twist measurements is what counts -- so the middle stripe is twisted off by 7 degrees relative to the bottom stripe. If the boom were in the picture, we'd be able to get a twist measurement for that, and so see how much the bottom stripe is twisted relative to the boom.

The camber measurements are pretty much what I'd expected for this sail, which should have about 8% max draft at boom, 9% at bottom draft stripe, 10% at middle, and 10.5% at top. The position of maximum draft is a little forward, though, at 32-38% of chord; I'd have preferred to see something between 40-45%.

Using the UK Sailmakers software is simplicity itself. Fire it up, open your digital sail photo, and click on one of the measurement buttons on the tool bar. Three buttons -- draft measurements -- are provided, called "Top", "Middle", and "Bottom", or "T", "M", and "B", but they can be placed anywhere over the photo. You can see in the screen shot that the "M" and "B" measurements buttons have been clicked.

When a measurement button is clicked, the measurement spline is placed on the photo, ready for use, and its results window pops up. Drag one end of the spline to an edge of the draft stripe on the photo, drag the other end of the measurement spline to the other edge of the draft stripe, and then drag the central "o" until the spline shape matches the draft shape. You're done! (Well, you can play with the other "o" control points of the spline to match it better to the draft stripe. You can also unclick the "Spline on/off" radio button, which leaves the spline curve on the screen, but removes the five "o" control points. With the spline control points "off", you get a tangent line instead, making it much easier to position the spline curve to the point of exact maximum draft.) The results are reported in the window.

One thing to note. Because the software is compatible with Windows 3.1, your file names should be restricted to 8 characters. If you use the long file names permitted by Windows 95 and 98, you'll find them truncated to "xxxxxx~1" and so on, which can make opening the right sail photo tricky...

So now you need some sail photos. You can take your usual 35mm (or whatever) conventional snaps, and scan them in for measurement. You can use your digital camera, and download your stills into the PC. Or you can use your digital camcorder with frame capture software. I thought I'd need to make a custom camera holder to keep the distance between camera and hull fixed from session to session. It turned out that my very ordinary tripod does the job perfectly -- set everything up to its minimum position, put two legs over the boat stand, and it is pretty much locked into a standard, repeatable position every time.

Camera in standard position

If you scan in your 35mm snaps, you do not need to scan at a very high resolution. 300dpi is overkill; 150dpi or even 75dpi is fine. Save the resulting scan as a JPEG rather than a BMP, and you'll end up with a file that should be no bigger than about 30kb. Nice and small.

If you use a digital camera (well, this is the best reason you'll find for buying one), you don't need mega-pixel resolution. My real cheapo Epson QX10 gives very acceptable results at 320 x 240. My relatively cheapo mini-DV Panasonic camcorder gives frames at 696 x 540 which are brilliant.

Larry notes that the software does nothing really new. It only makes easier what top sailors have been doing for a long time, and more sailors should do.

Larry also notes that there is a small error in the reported results because of the camera angle that is necessary for taking pictures of our model yacht sails. On big boats, you can get right under the sail, but for us (certainly when close-hauled) this is not feasible. I've put together a little spreadsheet (now with revised and improved formulae for entry and exit angles, about 26 kb) that corrects the reported measures for camera angle. Put in the distance of the camera from the mast and deck, tell it how high the draft stripes are, and it calculates the adjustments.

The spreadsheet, currently at version 3, also does a calculation of the entry (and exit) angle at each stripe, using the draft at 15% (and 75%) as reported by "Accumeasure" to make its estimates.

If you're interested in how the spreadsheet approaches this, it first makes an estimate of the entry (and exit) angles based on the amount and position of maximum draft.  These first estimates are remarkably good, but can be improved by using the drafts reported at 15% and 75% of chord.  In order to make "improved" estimates, the spreadsheet calculates what the 15% and 75% drafts would be if its first estimates were correct.  The first estimates are not correct, of course, and the spreadsheet then calculates what the maximum draft would need to be in order to result in the values actually reported by the Accumeasure software.  Using this, it estimates improved entry and exit angles.  Calculating what the max draft would need to to be involves a linear interpolation that uses a multiple of the discrepancy between the first estimates and the actual drafts at 15% and 75% of chord.  You can change this multiplier in order to improve the new estimates if you like.

2010-04-05


2011 Lester Gilbert