Home | Articles | Links | Projects | Columns | Designs | Boat Index

Questing Beast
Design by Milton "Skip" Johnson skipj@argolink.net

Essential statistics

LOA    21 '-4"
LWL    20'-2"
Length trailering    26'+/-- tip of rudder to hitch
Beam (hull)    6'-0"
Beam(waterline)    4'-8"
Beam(trailering)    84"
Beam (overall)    12" 2"
Draft (hull)    8"
Draft    20degrees heel 1'-8"
Interior Headroom @ berth flats    3'-2"
Design Displacement    1500#
Trailering weight    900#(+ trailer)
Sail Area    160 square feet
Mast Height AWL    20'-8"

Description

QBI is named for the Questing Beast from the novel. The Once and Future King, in which the Questing Beast was the lifelong obsession of King Pellinore a knight of the round table. They shared innocuous relationship that defined their very being, but was of no importance to either society, or to the events of the day. I've always been struck by how appropriate the name would be for one of my boats. My only previous use of the name was for a Mylar covered kayak named the Leasty Beastie during its short, UV ravaged existence. Most of my boats have more generic names like 14' solo canoe or more descriptive names like Easy Does It (16' Kayak) or Bionic Log, a wood veneer covered foam sit on top that quickly became Bionic Waterlogged.

The Rig, Sail and Foils

The rig, sail and lee foils are both the best and the worst (less tested) aspects of the design.  There won’t be much more development until spring. I'm a warm water wuss, and my friend Brendan Conroy always finds himself very busy during the holiday season.  Let me point out right now that it is my firm opinion that there is nothing so elegantly simple as a double blade paddle and a boat to go with it. The canoe additions are just a way to try out some concepts without investing a lot of time and money. The lee foils are about 12% thick ogival (circular) sections made out of salvaged door skin ply connected with a curved crossbeam laminated out of 6mm construction (cheap) plywood on each side of a piece of 3/4" blue foam insulation board. The lee foils are bolted to the crossbeam and the crossbeam is strapped to the canoe through the spaced gunnels. The mast is a 12' piece of 2" 0.60 +/- wall aluminum tubing left over from a previous attempt at the same general concept 15 or so years ago. The sail is a piece of Tyvek house wrap some 60 square feet in area with PVC pipe battens and a curved, laminated boom that rolls on a surplus $1.50 plastic square cross section pulley. The sail is biaxial. That is, it is symmetrical about a vertical centerline and flips (rolls) from one side of the mast to the other on different tacks. In my first experiments some years ago on a previous canoe the sail had two tacks, and like Phil Bolgers Proa sail, it was a mess that was compounded by the circular plan form and angled lee foils that ventilated like crazy. That early rig worked like gangbusters in bursts followed by tangles, twists and Chaos.

In the first of two and also the only outings with the current rig, the following was discovered. Surprisingly, the thing that worked least well the first time out was the rudder. My hold down setup for the rudder was terribly inadequate, and the Dacron ‘cables’ and foot bar weren't much better. We did determine the lead of the sail plan to the lee foils was far too much, since the canoe had an enormous amount of lee helm (possibly aggravated by my 190# being too far back in the boat. The lee foils did seem to work just as expected, providing just enough stability and not obviously ventilating when pressed. With all the lee helm, it was hard to determine much about the windward ability of the rig, but downwind was plenty quick enough, particularly with a rudder that didn't work and a stone lined bank rapidly approaching.

Back home, the crossbeam was turned 180 degrees to move the mast back about 12" and a slightly better rudder pedal setup was installed which also moved me forward about 8". Back to the lake. Much better. The rudder stayed down, and stepping on a pedal caused the boat to turn. The-sail rig is still going to take some work, but I was tacking through about 100 degrees, discounting all the way I lost coming about. The big problem in coming about (after thinking it through after the fact), was that the sail was going way out over the bow when swapping sides which kept the boat pushed over on the current tack. It is also quite possible to go over 90 degrees if the roller jams a little bit and you get the wind on the wrong side of the sail and pinned against the mast. All in all, the first two trials were great fun, and I hope to carry on with the trials a bit more in the spring before taking on the bigger boat.

Advantages and disadvantages. For the lee foils, given the need to run (and beat) in shallow water, I can't think of a better system. A low aspect ratio foil is always going to be less efficient than a higher aspect ratio, but in 2' of water it’s best to do the most with what you've got. Rectangular planform with endplates gets the most from the available depth. Actually, there's a little reverse taper to counteract ventilation on the inside surface of the foil. Running upright or under power(ed), the foils are just out of the water and not adding any surface area drag.

The sail is a little more problematical, many advantages, but untested. Advantages include, efficient foil shape, with no blanketing by mast. Easily reefed, with no change in balance. No chance of a jibe, plus running downwind has a slight upward lift component to help keep bow up. End(s) of boom are automatically far from the water and not likely to trip under any circumstance. The elliptical planform is very efficient almost by definition. The balanced layout allows a large area without requiring winches or multipart purchases. With practice, it should be possible to back down and maneuver in ways not possible with a conventional rig. Disadvantages: it is unconventional. Tacking will require handling at least two different lines and quickly to boot. Its also possible to get in trouble with the sail backwinded when there isn't any maneuver room. There is a lot of line in the design.

In short, to my admittedly biased eye, it looks really good, but different.

QB I is basically an oversized canoe fitted with a Birdwatcher style superstructure. I realize that most entrants in the contest will use stitch and glue type construction, but wood stripped canoes are the only way I know to build. The extra effort is minimal, confined to the hull proper and truly beautiful when done. Actually there would be a bit of stitch and glue at the juncture of the "(12mm) window frame and the deck. I had agonized over how to build that part without a lot of elaborate temporary framing when I realized that two precut pieces at 90 degrees to each other would fall right into place.

Construction

The particulars of construction are pretty conventional, 5 permanent bulkheads of 9mm ply, pre-finished, plus temporary frames of 5/8":particle board or equivalent at 16" o.c. provide the form for 3/8" x l" cedar strips. Once faired, outside of hull is covered with 12 oz biaxial glass, 2 layers on bottom and stem. Turned over, inside is covered with one layer 8.5 oz. unidirectional across hull with epoxy/wood flour fillets at bulkheads and ends. Interior hull stripping to be left finished clear, balance of boat to be paint finish. Upright and chocked, fit in outriggers for lee foils, window frames and decks with epoxy and wood flour fillets. Syntactic foam, epoxy and microballons with perhaps some foam peanuts thrown in, form the bottom of the anchor well and porta potty base. Two part expanding foams have a tendency to disintegrate and absorb water over time so use at your own risk. Finish hull to suit

Foils

Lee foils are built from 6mm ply like a large model airplane wing. Pre-finish framing and interior of skins, then assemble. Pivots are based on 1/2" S.S. rods with braces fabricated from 7/8" S.S. tubing and standard bimini top hardware fittings. Strategically placed foam pads and heavy shock cord straps should keep the lee foils tucked in for trailering, and be quick and easy-to fold out at the boat ramp.

Mast

Mast to be built up from octagonal ‘birdmouth’ section, 6" diameter, heavily tapered, finished weight about 40#. That’s too heavy, but with roller at the bow and gallows, you should never have to handle the whole weight, 20# I can do. Based on my experience with a hollow kayak paddle, an octagonal 'birdmouth' section turns out surprisingly easy to build, if you prefab an assembly clamp (a board with a tapered hole in it). For the mast, make up several 12" square pieces from scrap particle board or whatever with properly sized tapered hole in center of square.   Gather several friends on a 20' long flat surface with all the stuff and assemble. An alternative would be a foam or very thin wall wood section with carbon fibre skin but I think there's enough new stuff here already.

Boom

Boom is laminated to radius shown or can be fabricated as a box beam. Roller assembly is shown to be Harken midrange traveller car with curved track. An alternative, at your own risk, would be a roller assembly based on some (3) in line skate wheels running in grooves routed in top and bottom of boom. Battens are made from PVC plumbing pipe cut to length with ends heated and flattened. Battens are curved bowstring style with deadeyes and cleats to adjust draft of sail. Sail to be fabricated from white polyethylene sail kit, with several extra grommet kits. Sail to be laced to battens with dacron line through grommets in sail. The area between bottom batten and boom to be gathered and darted as required to make curve, place grommets in darts.

Rudder

Rudder is fairly conventional, laminated up from several layers of plywood. Based on my kayak experience, the planform shown with a straight trailing edge and relatively thick foil section is much, much more effective than a flat plate. The 6" PVC center shaft for the rudder pivot is an affectation of mine; the effective area-of the pivot is around the perimeter so the center doesn't really do anything, feel free to disregard, but then you'll either lose the ladder or have to rig it some other way.

Electric Auxillary

The trolling motor and gel cell battery called for are primarily a loading ramp and dockside maneuvering device rather than a true auxiliary. The units specified should manage 4 or more hours of operation on a full charge, running at almost hull speed, which should be sufficient for the intended use. The batteries also help pay their somewhat expensive way by serving as some functional self righting ballast and promising not to drip acid when upset. A solar panel would probably not be very cost effective, since QBI would normally be under cover except when in the water.

With about 60 square feet of wetted surface in the hull and a rig that holds its shape in a calm, QBI should ghost with the best. If there's absolutely no wind, go for a swim, it's relatively easy to board with the concentric ladder at the rudder.

Milton "Skip" Johnson
Home (281)373-0815
13022 Maxwell Road
Cyprus TX 77429
Office (281) 890-3893
Broad Spectrum Services
10601 Grant Road, Suite 110
Houston, TX 77070

Email: skipj@argolink.net

I'm a 57 year old Architect (small commercial & residential) with a life long love of boats, all kinds except maybe jet skis. Built my first boat 40+years ago, a 15' canoe.  Currently have a Larsen 16' Flyer, which replaced the runabout we'd had for 20 years Home built boats currently in the stable, 14' solo canoe, 16'kayak and my granddaughters' 8' first canoe, all strip built.

Drawings:

Sail Plan

main.gif (18766 bytes)

Hull

topandside.gif (41657 bytes)

Sections

sections.gif (23277 bytes)

line.gif (878 bytes)

 

Home | Articles | Links | Projects | Columns | Designs | Boat Index