Back in the 1990’s I had an idea to design and build a small proa for camp cruising on the Salish Sea. In part, I was inspired by L. Francis Herreshoff’s description of minimal yet elegant cruising aboard Rozinante, his whaleboat type, canoe stern ketch that he outlines in ‘The Compleat Cruiser’.
The other part was inspired by Hobie Alter and Rudy Choy, and all the Watermen of the Pacific. I said to myself, “Why not a sleek and un-ballasted version of Rozinante?” A little gem of a multihull for inland passages, gunk-holing, and exploring.
And this is where I went seriously wrong, because, in the back of my head, Bernard Smith, the author of ‘The 40-Knot Sailboat’, was also speaking. I began to see the project as not only all of the above, but also a research project, a sort of experiment into sailing dynamics that might result in a breakthrough!
Bernard Smith was a rocket scientist employed by the U.S. Navy, who happened to be fascinated by sailing dynamics and the creation of the ideal sailboat - capable of 40 knots over the open ocean - that he called the ‘aerohydrofoil’. Though Bernard was used to virtually endless resources to fund his ICBM research, he relied on his own funds for aerohydrofoil R&D, and Bernard mostly stuck to models, realizing that full-sized craft would quickly outstrip his personal finances.
Not me. I figured I’d get it right on the first try!
To make a very long story short, I didn’t get it right. Rozinante barely sailed at all, let alone well. It was a total disaster, and it eventually became obvious that to make it right would require a lot more cash than I could afford. The proa was a lame duck, a white elephant, and eventually all I could do was give it away to another unsuspecting dreamer (proa resale value is measured in negative $).
As this thread continues, I intend to relive all the missteps and wrong turns of the Rozinante project, mostly in the hope that my experience may be a dire warning to others. Stay tuned!
Michael,
I look forward to hearing more about what went wrong with Roz. By all appearances in the images, she was a very pretty boat with no obvious depatures from basic Proa design. The whole concept/size is basically where I want to go with a design….
Tom
this seems to be a perfect place to introduce myself, g’day, i’m Doogs.
i love sailing more than about just anything else, light fast boats especially. innovation, i love that too.
ergo, proas, those who build and discuss them have me intrigued. i look fwd to further installments on Rozinante…
Hi Tom and Doogs, and anyone else reading! Welcome to my nightmare.
The attached drawings show Rozinante as designed and built. Particulars are:
LOA: 26’ (7.9m)
BOA: 20’ (6.1m)
Beam hull CL to ama CL: 14’ (4.3m)
SA: 300 sq. ft. (27.9 sq. m)
Displacement: 1000 lb. (454 kg)
Construction: wood/epoxy, using the Cylinder Mold technique.
The basic form factor is fairly conventional as far as proas go. Steering was via “dagger rudders” (rudders in daggerboard trunks so they can be raised). The rudders were symmetrical ogive sections, so that they would not require reversing during a shunt. Both rudders were intended to remain immersed at all times, the forward tiller being locked into place so that it acted like a daggerboard, while the helmsman steered via the aft rudder.
The lee pod was there for emergency righting moment and it created a nice berth flat. I made a “pop-top” deck that hinged open to provide good interior volume while at anchor, even 7’ headroom!
All well and good, but what really set Roz apart was her rig. After reading and pondering muchly, I decided that the ultimate proa rig would be one that took advantage of the inherent bilateral asymmetry of the proa geometry, while respecting fore and aft as mirror images. You can see that line of thinking in the rudders, and I followed it with the rig.
Rozinante was designed with a high aspect ratio “square sail”. The sail is symmetrical fore and aft, but has permanently curved battens to give the sail an airfoil shape. Since a proa never tacks, the big disadvantage of the square rig (wearing around) is removed, and in fact it becomes an advantage because the rig does not require reversing as do all proa rigs. A few degrees of adjustment, the leech becomes the luff and voilà! We sail off on the new tack - a shunt that would probably re-align your vertebrae with the acceleration! Pretty brilliant.
I didn’t invent it. Bernard Smith used an ogive airfoil on his aerohydrofoil. The AYRS members came up with something similar that was published in their amazing book ‘Sailing Hydrofoils’, and they thought so highly of it they even called it an AYRS sail! Philip C. Bolger published a ‘cartoon’ of a proa with a similar rig in the gone but not forgotten ‘Small Boat Journal’, though with his own innovations.
The rig has much to recommend it and you can read all about it in the Proafile Archives. I was more or less convinced the rig was the holy grail of proa rigs, So Rozinante would have it. And this was my hugest mistake, I just went and built her, without taking the time to build a scale model or a test boat or anything. I honestly thought that the rig could not fail!
Picture the first sail: A beautiful late summer’s day, north wind of 12-15 knots, whitecaps on Lake Washington. Ideal Hobie weather, as we used to say. The boat is towed out from my parent’s dock using the venerable Sea Ray as tender. The “paparazzi” are aboard, camcorder aimed at the new boat and crew, which was myself and my brother.
We get the sail up, trim in, and the boat is off like a rocket on a reach! Yeeeee ha! Until… suddenly the sail slams into reverse gear, acting like it was suddenly back winded! And then immediately it would slam back out again, pulling us along like a freight train. Yikes. Slam-stop! slam-go! slam-stop!!! Finally, one of the crazy slams heeled us over so that the lee “safety” pod dove underwater. Which promptly filled with water since the pop-top was not water tight. Oops! That suddenly added many pounds of ballast in exactly the place you don’t want. Aack!! For reasons known only by Poseidon, Roz did indeed right herself, and we eventually made it back to the dock, tail between our legs.
Next Installment: aftermath.
Wow,
I’ve read about the problems/dangers of that rig. I’d always wondered if it could be tamed somewhat by using rigid straight battens and building the curvature into the panels like a modern junk. That way it would maybe be able to luff/depower to some extent. Still, when shunting there would be a point where the wind was at 90 degrees to the sail as the leading edges swap…....
Bolger actually showed heading up to shunt, maybe for this very reason.
Looking forward to “The rest of the story” 😉
Tom
So what was happening? Why was it powering up and then backwinding?
I had a few more outings on Rozinante after that epic first sail, but even though I had designed in plenty of room for adjustment, the sail was still a monster. With fall setting in, I took the boat apart and stored it beside my brother’s contractor’s shop, and there she sat. I had no idea what was wrong, and no idea what to do to fix the problem.
A year or so later I made the acquaintance of John Dalziel, who was testing proas in Florida. He’d also experimented with a similar sail and had experienced similar frustrations, which he wrote about in his piece on the Bolger proa rig. To be perfectly fair, the rig I tested was not exactly a Bolger rig, but I hadn’t set out to copy Bolger. My ideas were influenced by the AYRS and it was merely coincidence that Bolger happened to publish a cartoon of a proa rig that was uncannily similar to mine. In hindsight, I think the strong roach built into the Rozinante sail was at least partly to blame for the crazy behavior. After all, the sail was, more or less, simply a large, balanced club jib - like they used to use on ice boats.
Another design error was that the sail was canted to windward in an attempt to gain aerodynamic lift. Trouble was, the sail, with it’s rigid aluminum battens, was damn heavy, and it took a bit of wind to even get it to set! So forget light air performance. Also, the geometry of a windward canted rig means that as the boat heels, more, not less, sail area is presented to the wind. This is exactly the opposite of what is desired.
A few years later, I decided to build a scale model of Roz and do the testing that I should have done at the very beginning. You can read all about that in Testing With Models 1 and Testing With Models 2. I learned an amazing amount from doing that, I can’t recommend scale model testing highly enough!
Anyway, eventually I got back to work on Roz. Even though I wasn’t sure what to do about the rig, I knew the lee pod needed to be water tight so I built a nice little cedar strip blister cabin for her - no more pop top. I also had some rot issues to deal with after her outside storage at the shop. Soon after that I moved to Whidbey Island, my brother closed his shop, and I took Roz along with me. A few more years passed and eventually it became clear that Roz just wasn’t going to fit into my life anymore, even if she had the most awesome rig in the world. I’d moved on, and she was holding me back. All I could do was give her away, since she had no value whatsoever on the used boat market. I was thankful I found someone who was willing to take her.
These photos are from Whidbey Island, summer of 2000, the last time I had her assembled.
Next: what I would have done differently, if I had only known.
What an attractive boat, if I had not been 1000’s mile away I certainly would have taken her.
Interesting about the rig. Was the sail moved back on the mast for reach tack?
I see James Brett appears to use something similar with success.
Brett’s rig is actually very close to a traditional junk. instead of curving battens, they’re straight and the camber is cut into the sail between the battens so it will actually “luff”. It also keeps the same halyard and downhaul with the sail rotating around the mast in a traditional manner. He’s got the traditional junk sheeting that controls most of the battens at the leach avoid excessie twist, though it looks like he’s got it set up to “shunt” the beckett block from end to end.
If you go to Gary’s sight and look at his junk Wa’apa, it’s very close to the same setup. This is actually MY preferred rig.
Tom
The rigs are superficially similar but that’s where it ends. Back in the 90’s, conventional wisdom was that a junk rig was too inefficient for a multihull, so I didn’t even consider it. I’m happy to see Brett and others take on the junk and prove the conventional wisdom wrong. Most junks I’ve seen have unstayed rigs, which I like a lot but Roz was designed for a stayed rig, to take advantage of the wide staying base. I’m pretty sure you could still use a stayed junk rig on a proa. It would have been a good option for Roz.
Cambered Panels have made a world of difference for modern junk sails. Apparently we had to re-learn that. The original Chinese sails were cut flat, but ended up with camber because of the stretch in the material they used. Us bright westerners “tidied up” those baggy sails with modern materials, and all of the sudded had trouble going to windward. People were trying bendy battens, hinged battens, all with the problem of obtaining more curve the harder the wind blew, the OPPOSITE of what’s needed.
This link is to my first Youtube attempt showing my first junk rig attempt. My mast is too long, my halyard too far forward on the yard (which is too bendy), too much tension on the down haul (the boom is too bendy as well), and the tarp material I used is too heavy but it’s still working well.
http://www.youtube.com/user/tdsoren1?feature=mhee
I’ve also got photos of the original and modified rig at http://groups.yahoo.com/group/junkrig/photos/album/775420805/pic/list
I’ve since done a lot of modification short of actually sewing a new sail, but the weather has turned so it will be a bit before I get back out for more video. The beauty of this rig is the downwind manners and just how fast and easy you can reef, allowing you to really push on the sail area.
Tom
What a nightmare that first sail must have been, Michael. Full marks for sharing the experience. Looking forward to hearing about the lessons learnt!
Tom: thanks for the info and links to your junk rigged tri. I am a HUGE fan of the junk rig. It’s beautiful, easily handled and also very DIY friendly. Please keep us posted on how it goes next year.
James: Yea, for some reason I feel like I need to share the experience in order to properly move on.
So after all that, what would I do differently, if I had the chance to do it all over again?
Quick reply: go look at Madness - John Harris has basically designed Rozinante II for me.
But to make a longer reply, I thought I’d make a few rules that, if I had followed them, would have saved me a lot of grief:
1. Don’t make a 26’ boat your first boatbuilding project. Start small, make small mistakes.
2. Experiment with scale models first, then a one man test platform, then finally at full size.
3. Stay clear about your goals. Do you want a cruiser or a speed machine; a boat that will work “out of the box”, or something experimental?
4. Think about auxiliary power. Proas are not easy in that regard.
5. Think hard about demounting, trailering, storage and moorage. The boat needs to fit into the real world and that includes a lot of transporting from here to there.
6. Design for your technical boatbuilding abilities, and be honest.
7. Design for as much DIY as possible, this includes spars and sails.
After all the above, I want to say that I’m still pretty jazzed about the potential of a “square sail” rig for a proa. I didn’t get it right, but I think it will be figured out, and when it finally is, then the modern proa will really be ready for prime time. I’m really excited about Skip’s new proa design - ‘Nomad’.
Hi Michael,
what a story. Thanks for sharing. I think I saw the first drawing of Roz in the portfolio of Kurt Hughes, is this correct? It made me think about proas the first time. A team of finish students have built and sailed it. But I dont know how good it sailed.
I feel very sorry that you did not found the power to continue with another kind of rig and other modifications.
I want to make a few comments to share a little of my experiences I made last year.
A few additional comments about modell testing.
My last winter project was the design of a 26’ sharpie and i began to build it as a rc model in 1:10 scale (your common 1’ = 1’’ is 1:12). To learn the needed know how I entered a forum and met very friendly and experienced model makers. They told me that the boat was far to small to give reliable results as a scale model representing a real boat and in my case I should use a 1:5 scale to get a model 1,6m long. That would be a length that will work. Of course I finished my 0.8m Model and went sailing, which was a great fun. But I needed more ballast for proper sailing.
See it sail here http://www.youtube.com/user/luckystrike118?feature=mhee#p/a/u/0/05Co29RgZDc
The fact is that you can scale down static stability like 90° righting moments für capsise calculations, but you cannot scale down sailing stability and wind forces.
Later I learned that especially multihull models have the problem that the density of water is not scaled either, so the modelboats are sailing thru oil, to say it figurative . That means ... if your modell has a slight bow down attitude and pitchpoles when it is pressed very hard, it must not nessesarily behave like this on the real boat.
Modeltesting is very attractive and reliable, but make the model at least 4 feet long, better 5’. Your results will be nuch nearer to reality.
The fastest method to make functional scale models is the use of styrodur foam, cut by hotwire into the correct profile and then shaped like a custom surfboard. Use templates for deck outline (and bottom Panel if you have one) and main bulkheads. This way you are ready for laminating with epoxy in one or two days. 2 layers of 4 - 5 ounce (flexible) woven glass will do.
Have Fun with it, Michel
I traded some illustration work for engineering work with Kurt. He worked out building Roz with his cylinder molding technique. He put the design in his portfolio after that. Thanks for the model building tips.