I once made a cockpit awning. It was a fiberglass-over-plywood affair. Not only was it a cockpit awning, but it also could have been pressed into service as a mediocre paddleboard, a bus shelter for small children and/or midgets, a roof for a tiny gazebo, a protest sign, a miniature frog pond and, of course, a planter. It turned out to be a universally useful/useless piece of crap, depending on how you looked at it.
It started well. I used 1/16-inch Luan for the top and narrow slats of 1/2-inch for the frame, which I cut to gentle curves that made the top into a cold-molded conic section with just a tiny bit of spherical distortion for added stiffness. I filleted the inside joints, sealed the plywood with epoxy, fiberglassed and faired the top… and then I tossed it. Actually, I gave it to some artists, thinking they might use it for some sort of art installation. It didn’t make that good a cockpit awning: too heavy, too difficult to mount securely, plus it added too much windage aft. I didn’t think it would survive a hurricane (unlike the hard dodger I made earlier, which survived passing close to the eye of Hurricane Matthew with no damage).
I did most of the work on sawhorses on the floating dock at the marina. All of the other marina denizens, who mostly just sat on their boats and got drunk, were rather enthusiastic, and a few even tossed some business my way, fixing stuff on their boats. But the marina staff were less enthusiastic, talked about made-up “customer complaints” and eventually exiled me, together with my sawhorses and tools, to a windless, gravel-paved back lot, where I worked roasting in the sun. The hostile work environment probably had something to do with the project’s ultimate failure, but mostly I blame myself, for not spending enough time on the design phase.
There are plenty of designs that are specifically useful for their stated purpose, but are otherwise completely useless. In this category are special-purpose tools, like the egg slicer or the lemon juicer. Yes, they make short work of slicing hard-boiled eggs or juicing lemons, but beyond that they just add clutter. In a pinch, both can be used to prop open doors and windows, and the egg slicer makes a tiny out-of-tune harp, in case you are ever in need of a really pathetic sound effect. But that’s it.
Lots of boat designs are the same way. Most yachts, for instance, are only useful for showing off how rich the owner is (or was before he bought the yacht). Sporty ones are only good for pounding across the seas slightly faster than the competition and in great discomfort. Shantyboats, sailing scows and single-wides floating on barges or pontoons are cheap to maintain and comfortable to live aboard, but they give harbormasters and marina managers the vapors because they don’t look sufficiently yacht-like.
In setting out to design Quidnon, my objective was to create something sufficiently versatile to make it one’s single most valuable possession. It is a houseboat, a motorboat, a sailboat, a party boat and a beach house. It can handle deep water as well as the shallows. This level of versatility calls for some amount of compromise, and the question is, How much compromise is too much? “Better is the enemy of good enough” is a good saying, but how does one go about determine what’s good enough? And when should the alarm go off to indicate that a design has cross the line between generally useful and generally useless?
This requires lots of careful thought, and that is why the design phase of the Quidnon project is taking a few years rather than a few months. A typical boat for a rich guy to show off on or for a charter fleet is relatively easy to design. The design of a very unusual boat that will be useful as a home and a magic carpet to many different kinds of people all over the planet is far more challenging. And yet I think we’ve come quite far. A dozen or so well-defined design tasks stand between us and a set of plans from which we can build the first hull.
I know that are number of you are waiting for that moment. I am sorry to make you wait, and to make up for that somewhat I want to share with you the complete list of tasks to be completed before we can produce the design plans. Very importantly, these tasks need to be thought through before they can be drawn. As they say in the world of software, “with enough eyes, all bugs are shallow.” So far, this has been the case with Quidnon as well: over time, good ideas have been added and bad ideas eliminated simply through knowledgeable, thoughtful people asking good questions. If a question doesn’t get asked, a bad idea can stick around for a long time.
A case in point: Willie, a marine engineer who recently joined the project, asked a simple question: Why are there twin rudder blades? The boat doesn’t heel much, so angling the rudder blades doesn’t add much to the performance. The Ackermann steering geometry requires a complex (and expensive) linkage. And accommodating the twin rudders complicates the hull shape at the transom. Why not have one rudder, located amidships, where it can deflect the prop wash, for better maneuverability?
Some of this logic also applies to the twin keelboards. If the boat only heels 12º even when pressed hard (as shown by the scale model) then angling the boards has little benefit. But the second board does add complexity and cost; why not get rid of it and just have a single centerboard? It can be mounted off-center, as in some of Phil Bolger’s designs, to keep the center of the cabin unobstructed by the centerboard trunk.
And the answer is, I initially added the twin keelboards and rudders before I knew how little Quidnon would heel, and I didn’t revisit the question because until recently nobody had asked it. So, please ask! With that in mind, here are the known tasks to work out.
• Add a small deck arch at the bow to serve as a mast support
When the masts are taken down, they fit within the overall length of the boat while sitting on the deck arches with the sails (along with spars and battens) slung below them. But the masts are unsupported at the bow. Adding a post at the bow would preclude a boarding ladder from being deployed off the bow. This is useful when docking bow-to (to pick up or drop off passengers quickly), when nosing up on a beech or to an ice floe, etc. The arch should be skinny so as to not obstruct the view forward. And it should not incorporate vents, as do the other two deck arches, because there is often too much spray at the bow that would make it through. Ventilation for the U-berth will need to be provided in some other way, such as through airducts run under the cabin sole.
UPDATE: In response to this, Matt suggested that the mast tabernacles incorporate shelves to support the masts. This is a very good idea. The foremast is close enough to the bow so that this change would make no difference. Mast shelves on the mainmast would be useful too: when the mainmast is first lowered down and the tabernacle hinge pin removed, the forward end of the mast needs to rest on something before it can be pulled forward. The mast shelves should extend aft of the mast tabernacle hinges so the masts can rest on them as soon as they are unhinged.
• Add dinghy forks aft
This is a relatively small detail, but important. There is ample room for storing multiple dinghies on deck, but it is often helpful to be able to deploy a dinghy quickly. Setting a dinghy upside-down on dinghy forks that slide out from the transom and lashing it down securely is in many ways optimal, and in my experience better than hanging the dinghy from dinghy davits so that it rocks, accumulates spray and rainwater and blocks the view aft. The dinghy forks can be used as dinghy davits when Quidnon is at anchor or at the dock, just to lift it out of the water, to keep it from accumulating marine growth and to give would-be dinghy thieves second thoughts.
• Add skids to the bottoms of keelboard trunks
Having straight skids is useful in a number of cases, such as rolling the boat ashore over round sticks and dragging it onto and across ice. If one of the keelboards is eliminated, there will still be two longitudinal full bulkheads that can be extended below the bottom to form the skids. The bottoms of the skids will need to be fiberglassed heavily and finished with epoxy thickened with graphite powder to provide a durable, low-friction surface.
• Finalize design of sliding doors
There are a few places where two-panel “Star Trek” sliding doors (minus the silly swish-swish noise) make a lot of sense. We already have a good design that uses counterweights on loops of cable to keep the panels from sliding open or closed as the boat rocks. It just needs a couple of tweaks. The main one is to have two counterweights—top and bottom—to compensate for angular momentum.
• Design stoves for heads and galley
The two stoves can be identical. They need to be able to burn propane, wood or charcoal. When burning propane, burners are inserted into what is normally the ash pan; the firebox can then be used for baking or broiling. The top of the stove is a cooking surface for the galley stove and a rock heating surface (to make steam for the sauna) in the heads. There need to be two thermostatically controlled louvers to divert flue gas flow to two heat exchangers. One heat exchanger is air-to-air, the other is air-to-water. The hot air is piped around through ducts under the cabin sole and to the cockpit well, for heating. The hot water is piped through an insulated hot water tank, to be used for washing and bathing. In freezing weather, some of the hot water needs to be piped to the water ballast tanks, to keep them from freezing.
• Rework joinery to use “screw and glue” rather than “mortise and tenon”
The scale model, and the earlier plan, used a lot of mortise and tenon joinery. It worked quite well in some places and less well in others. Specifically, it worked well for orthogonal joints and badly for joining elements on a curve. And it suffered from three major overall defects: 1. because the joints had to be kept a bit sloppy to make assembly possible, it soaked up a lot of epoxy, adding weight and expense; 2. it turned out to be rather difficult to calculate the strength of these joints; and 3. a lot depended on how carefully the joints were filled with epoxy, leaving open the possibility of voids that would concentrate moisture and cause rot and in pinholes that would produce small leaks. For all of these reasons, we decided to use a much simpler joinery technique of using square or beveled fir sticks and screwing and gluing the plywood panels to them. This technique is time-tested, the pull strengths of fasteners and the holding power of epoxy joints are both well known, and the skill level required to achieve good results is quite low. But quite a bit of structural analysis needs to happen in order to determine the appropriate sizes and spacings of screws.
• Rework the shape of the bow and the transom
With the twin rudders gone, the shape of the transom is simplified. Previously, the bottom, where it meets the sides of the transom, had to be angled; now it can only be curved in one direction: fore-and-aft. The bow needs to be made deeper in order to compensate for the loss of some 3 tonnes of fixed ballast aft by adding a shallow stem to it, as I explained in a previous post. The addition of the stem will also help sweep aside floating debris and bits of thin ice. The exact shape of the bow will be determined by running Orca3D hydrostatic simulations, to make sure that the boat sits on its lines.
• Rework bow construction technique
This didn’t work out so well on the scale model because the curves are too tight to be cold-molded. I ended up having to steam-bend plywood, which is not something we should expect Quidnon assemblers to be comfortable doing. Plan B is to use a single layer of 1/8-inch plywood to create the shape, then use it as a male mold to lay up as much fiberglass as needed to give it the necessary stiffness and strength. On the other side of the 1/8-inch plywood will be a lattice of thicker plywood to support the shape from the inside.
• Rework sheer strip assembly, hull and deck joints
A major problem when assembling the scale model had to do with fitting the sheer strips, which were two layers of plywood. At least 3 layers of 1/2-inch plywood will be needed for the full-scale build. The holes for the deadlights didn’t line up and prevented the sheer strips from developing a smooth curve. It took a lot of clamps to keep it from becoming lumpy. So, the revised plan is to make the deadlight holes using a hole saw or a jig saw and a router post-assembly. Also, after a bit of math it turned out that the deck-to-sheer strip and sheer strip-to-topside joints needed reinforcement. The simplest way is to use perforated aluminum angles rolled to the right and curve and attached using stainless steel mechanical screws with fender washers and nuts. That’s a lot of hardware, but deck-to-hull joints are critical and notorious for developing problems.
• Rework the rudder to use a single, central rudder blade
The rudder blade assembly can be tucked under the transom into the recess between the engine well and the transom that is directly below the gas tank and the propane locker. The recess is already there so that the back of the engine well doesn’t catch waves or prop wash from the motor. The entire Ackermann linkage goes away (along with several thousand dollars’ worth of expensive hardware). Some amplification of the tiller angle using an adjustable linkage between it and the telescoping tiller extension may still be required to keep the useful swing range of the tiller inside the cockpit.
• Convert inboard sides of keelboard trunks into full-height, vertical, longitudinal bulkheads, then get rid of the port keelboard and its keelboard trunk.
This was a major area of concern. There is a lot of side force on the keelboard trunks from both the keelboards and the mass of the water ballast. The longitudinal bulkheads will have openings in them through which to access the pilot berths, which are on top of the ballast tanks, and the sides of these openings may need to be reinforced with vertical ribs.
• Create plumbing, electrical and air duct schematics
The plumbing schematic exists; the electrical schematic needs to be created. The routing for all of them needs to be laid out and components selected.
• Design engine mount
Similar engine mounts, in which the motor slides up and down instead of tilting, exist for catamarans, so it may be possible to repurpose or borrow an existing design.
• Complete design of standing and running rigging
The standing and running rigging for the sails needs to be tested on a physical prototype at 1:5 or 1:4 scale to work out where to place the blocks, etc. Of specific concern are the details of the mast parrels, the placement of halyard and downhaul for optimum sail tension, and the placement on boom and battens of sheets and reefing lines. Take-up spools for running rigging (which will live under the cockpit well, above the chain in the chain locker) need to be designed as well.
This is the list as it stands at the moment. A few more items will probably need to be added as we move along. If you have the time, the skills and the inclination to tackle some of these, please let me know; we are, of course, looking for more engineers to join the team. The work is on a volunteer basis until the project reaches the equity crowdfunding stage.
If any of this brings up questions in your mind, please ask! That’s the main purpose of this exercise—to see if anyone can poke holes in our plans, or open us up to ideas we haven’t thought of or considerations we haven’t been aware of.
The purpose of this project is to design and mass-produce kits for a floating tiny house that can sail. It combines high-tech modeling and fabrication and low-tech assembly that can be carried out DIY-style on a riverbank or a beach. This boat is a four-bedroom with a kitchen, a bathroom/sauna, a dining room and a living room. The deck is big enough to throw dance parties. It can be used as a river boat, a canal boat or even a beach house. It's rugged and stable enough to take out on the ocean. Kits will start at around $50k (USD). The design has been tested in simulation and prototype; full-scale production will begin next year.