On the shrouds page I theoretically uncovered rather small lateral mast deflections -- say, 2mm at the top of the mast, and 3mm in the mid-mast -- and thought that these were for my Ikon and might not apply to designs which did not have effective mast partners:
Here are two photos from some recent wind tunnel research undertaken by
Rachel Nicholls-Lee, a third-year ship science student at the University of
Southampton. Her work was sponsored by Graham Bantock and SAILSetc,
and with a little luck she and Graham will allow me to mention other interesting
In both photos, the model (a "Jazz" IOM) was aligned at an angle of 30 degrees to the apparent wind. In both, the main boom was sheeted at an angle of about 0 degrees, and the jib was sheeted at about 12 degrees (yes, very tight, but that was the aim of this part of the tests). The top photo is for an apparent wind speed of about 1.8 m/sec (yes, ignore the "2.0" on the photo), and the bottom for a wind speed of about 4.5 m/sec. The spreaders were "short" -- shorter than half beam -- and were located well below mid-mast, so the shrouds were pulled in somewhat. The shrouds were attached close to the jibstay (about 40 mm lower). Tensions were about 2.7 kg in the backstay, and 4.5 kg in the shrouds. The 11.1 mm mast had about 40 mm of pre-bend.
Although the mainsail took a nice shape at the lower wind speeds, by the time the wind reached 4.5 m/sec, the lower part of the mainsail began to develop an over-bend crease. Close-hauled, it was very modest, but as the wind freed the over-bend crease became increasingly significant.
Photo overlay (short spreaders below mid-mast)
When the two photos are overlaid, and the resulting image digitally enhanced, we can see what has happened to the mast and sails as the wind speed increased from bottom to top of "A" rig.
The mast position at the lower wind speed is shown as a thick red line, and at the higher wind speed it is a thick green line. Speaking generally, it is easy to see that the head of the mast has fallen off considerably -- perhaps 25 mm -- and that this fall-off, lateral mast bend, is concentrated in the upper mast above the spreader location. The mainsail shows considerable twist off at the head and mid-leech, partially due to the mast sag, and partially due to the light sail material yielding in the high wind.
Brian Hard has taken my photos and produced a much better overlay using AutoCAD. The following is his screen shot. It also makes the jibstay sag much clearer.
Photo overlay (spreaders above mid-mast)
Here is another digitally-enhanced image of two overlaid photos, but this has one very significant difference -- the same short spreaders have been moved up the mast to well above mid-mast. Instead of the spreaders pulling the shrouds in by being "short", they now push the shrouds out in the conventional way by being "long"; "long", that is, relative to their position on the mast. (There is another difference, in that the model is now angled at about 90 degrees to the apparent wind, and the main and jib are seriously over-sheeted at 40 degrees and 50 degrees respectively. This is not significant for the purposes of this discussion of lateral mast bend.)
Earlier during testing, on the beam reach, when the spreaders were in their usual position below mid-mast, the over-bend crease in the mainsail at the highest wind speed was totally excessive. In a "real" event, I would have been horrified to see this, and would have done everything possible to adjust the rig to remove it. So we moved the spreaders up the mast to regain acceptable mainsail shape.
With this pair of overlaid photos, the mast position at the lower speed is shown in yellow/gold, and at the higher wind speed in dark blue/black. Speaking generally, it is easy to see that the head of the mast has again fallen off considerably -- perhaps 20 mm this time -- but now there is very little lateral mast bend as such -- the whole mast has sagged to leeward. The mainsail shows the same sort of twist off at the head and mid-leech as it did previously.
Sketches from above
The top diagram, viewing the boat from above, is taken from the shrouds page, and shows what I thought would happen to a conventionally-rigged boat. With conventional, "long" spreaders, I thought that the mid-mast would sag to leeward (red curve), while the head stayed where it was or moved to windward. I was more or less right on the first assumption, but quite wrong on the second.
The second diagram shows what the wind tunnel shows -- a conventional "long" spreader arrangement keeps the mast relatively straight while the whole of the mast sags to leeward (purple curve). The green shaded components illustrate the lower wind speed, and the purple shaded items illustrate the higher wind speed.
It may be worth being clear what seems to be happening. The mid-mast has indeed sagged to leeward, and part of that is due to the conventional long spreader pushing the mid-mast to leeward. What is interesting is that the mast head has also sagged to leeward. It sags to leeward so much that, effectively, the whole mast remains relatively straight.
Sketches from behind
The diagrams below illustrate the same phenomena viewing the boat from behind. The left-hand diagram is a sketch based upon some test runs with the spreaders replaced in their "usual" position -- somewhere below mid-mast -- but lengthened to around 85% of beam. The right hand diagram is a sketch based on the "short" spreaders which were around 40% of beam.
In both sketches, the mast head has sagged to leeward. With conventional spreaders, the mid-mast has also sagged to leeward. With the short spreaders, the mid-mast has been much better supported, and is more or less straight. But the mast head sag introduces serious over-bend wrinkles into the mainsail when reaching. Perhaps a different, rather radical, mainsail luff curve could deal with this.
What can we conclude? First, of course, is to be rather more cautious in future when theorising in the absence of hard data!
We must note that the model was a Jazz and not an Ikon or Italiko. It did not have an effective partners clamp on the mast at a raised foredeck, and so the whole mast and rig was less rigid than would be found in these designs. The result is more mast bend and much more obvious twist off in the mainsail leech as wind speed increases. This is good, of course, from the point of view of seeing what is going on in the wind tunnel.
We must also note that the sails were "standard" SAILSetc "light weight" manufacture, made of light Mylar film. This meant that they were unable to resist a hard blow, and as the photos show, the mainsail leech twisted off dramatically at higher wind speeds.
The short spreaders did in fact partially work as theory suggests (see the "Spreaders" page). At lower wind speeds, they stopped the mast from sagging to leeward, and consequently kept the slot open. And the head did not sag off too badly and the mainsail kept its shape (keep in mind that the hounds were up at the jibstay attachment point on the mast).
The long spreaders also worked as theory suggests. They pushed the mid-mast to leeward.
It was at the highest wind speeds that my theory came apart. While the short spreaders continued to do an excellent job of keeping the mast from sagging to leeward, they could not prevent the head from bending off, and the resulting mast bend introduced unacceptable over-bend wrinkles in the lower mainsail shape. While the long spreaders continued to push the mast to leeward, this kept the mast more or less in column, with the mid-mast sag staying in line with the top mast sag, resulting in excellent mainsail shape retention.
Fore and aft mast bend
Finally, a note about what happens with fore and aft mast movement as the forces build up when the wind rises. The following overlays show long spreaders on the left, and short spreaders on the right. The model is on a beam reach. As before, we can see how the long spreaders have pushed the mid-mast to leeward, while the short spreaders have not. But we can also see that the mast has in addition moved forward mid-mast under wind pressure, because we can see two spreader images. The long spreaders are offset both towards the left and the bottom of the picture, while the short spreaders are offset only towards the bottom. In both cases, clearly, the mid-mast has moved forward. Looking at the head of the mast, this hasn't moved forward much at all, and has really only moved sideways. Conclusion? The jibstay is sagging very significantly.
A careful look at the left-hand image shows that, in this case, the mid-mast has been pushed to leeward, and the mast head is also to leeward, but not quite as much. Relative to the mid-mast, therefore, the mast head is in fact slightly to windward. The long spreaders, 85% of beam, are therefore probably too long.
A careful look at the right-hand image shows that the mid-mast has not sagged to leeward at all, because the spreader images show no offset to the left. But the mast head has clearly sagged off considerably, and so all of the lateral mast bend has occurred above the spreaders attachment point. This is what put horrendous over-bend wrinkles into the lower mainsail.
©2011 Lester Gilbert