Part 1 - Part 2 - Part 3 - Part 4

Introduction

It is often recommended, for first time boatbuilders, to start with a small project. That's one advice I followed: I built a one-sheet boat. Both the building and storage space available to me could not handle anything much bigger anyway.

Designing your own boat, however, is not recommended by experts. Well, since I am as much interested in the design process as in using that boat, I couldn't resist the challenge. My thinking was, this would be a great learning experience, and if the boat proved to be a failure, the investment in time and money would not have been too great. The project took 2 years to complete: one year to draw it, one year to build it. That took longer than expected, but I didn't fully know how to draw a boat when I started this project. In particular, I had to learn how to do a surface development for the hull. As for the building, the sanding took forever, it seems.

I am pleased with the way the boat looks. Performance wise, though, it is quite tippy, at least for a novice paddler like me. But it was a great learning experience.

History of the design

I did not wake up one day and just decided to build a one-sheet canoe, though. What happened was this: a few years ago, I started to play with Gregg Carlson's Hull Designer program. During the same period, there were quite a few one-sheet designs presented on the Duckworks web site. I thus chose to learn that program by concentrating on a one-sheet double-ender. Why a double-ender, do you ask? Well, first, this is an easier design to figure out. If the crew sit right in the middle, you get perfect trim. Since, as far as I know, the Hull Designer program doesn't adjust trim, this was easier to deal with than on a transom boat. Second, maximizing the length of the bottom and side panels seems easier if you don't have to also fit a transom on that single sheet. Third, I was aiming more at good gliding performance rather than the highest displacement possible, as with a pram shape.

The first hull looked like this:

Simple flat-bottom double-ender

This is essentially a flat-bottom hull with vertical top sides. That's about the simplest shape there is for a boat: all the curves are basically circle arcs. The reason for using vertical top sides without any flare was to achieve maximum waterline length. With some flare, the chine is typically shorter than the sheerline. The sheer curve is achieved by drawing a curved plank, rather than using a mostly straight plank with a good amount of flare.

Although such a hull shape doesn't look too bad, it is hard to achieve a fair displacement and adequate freeboard at the same time. I decided to go from a flat to a V-bottom, to see if this would increase displacement. I also started to make cardboard models of the most promising models, using empty cereal boxes.

Simple V-bottom hull

The hull shape did look a bit better, despite being drawn with vertical sides. The displacement was a bit better but still too low: my target was 250 pounds. More rocker in the bottom would have been necessary to keep the stem above the waterline. But this would have meant less freeboard amidships. As a novice designer, I wasn't sure how much freeboard was enough, but more seemed better.

I decided then to got for a second chine, hoping for better results. I tried several variations of plank shape for the new middle plank. The one variation that seemed to go somewhere was this one:

First noteworthy double chine-bottom hull

By making the middle plank narrower in the center and wider at the extremities, this left more plywood for the midship area of the bottom planks. This also flattened the keel line, burying the stem somewhat underwater. This would be bad for a flat-bottom boat, but good for a multi-chine hull. This improves tracking and adds a little more displacement. The shape was starting to look a lot like a canoe. It is at that point that I decided to seriously attempt to draw a canoe.

The middle plank was quite narrow on this model, looking more like it was just softening the edge of a hard chine hull. For the next model, I decided to use a wider middle plank. I looked more seriously at optimizing the midship cross-section. Hannu Vartalia's online essays, in particular, come to mind. I also checked what were typical dimensions for a canoe. An online version of W.P. Stephens' book 'Canoe and Boat Building' proved useful for that. It might have best to go look at real canoes, though. I might have noticed that typical canoes have a flatter midship section than what I was drawing. Anyway, the next model looked this:

Double Chine V-bottom, wider middle plank

All that was missing for an authentic canoe look on this model was a round stem/stern and a round bilge. The round bilge couldn't be done in plywood, as far as construction was concerned. The round stem, I couldn't do using Hull Designer. You can approximate the rounded look by creating multiple chines, but that program, as far as I know, is not suited for drawing perfectly round stems (you can't ask too much from a free program). At that point I moved to another program for drawing my hull models. I used TurboCAD, a general purpose drafting program I was somewhat familiar with. I say somewhat, because although that program can handle drawings in 3D, I was not comfortable with the 3D interface of the program. Instead, I drafted my next hulls in 2D, much like the way hulls lines are drawn on a sheet of paper. This meant much slower progress, having to deal with surface development manually. One book, by S.S. Rabl, titled "Ship and Aircraft Fairing and Development" proved helpful to understand the process, which previously was a mystery for me. It nevertheless took several drafts, due to various mistakes and problems, before I could build a cardboard model.

3rd draft using TurboCAD, the first attempt complete enough to make a model. Baby Canoe is now the official name of the design.

The round stem gave me some trouble: a surface can be developed if it bends in one direction only. With a hull shape created by an arc lengthwise, near the bow the plywood would have to bend both horizontally and vertically: not possible. The solution was to use an arc for most of the boat length, but finish the last 9 inches near each end with a straight line. This way the surface would end up closer to what plywood can handle: it would only have to bend vertically to follow the curved stem.

The stem curve was achieved by using a portion of an oval shape, with a straight section for the top plank. That curve looked fine in profile, but the lower part of the bow seemed too full, especially when looking at the model. If you look closely at the hull lines, in the profile view, the bottom chine is almost a straight line. It didn't look too bad, but it certainly looked odd in profile view. I thought it would be best to raise that chine near the bow and stern. A couple drafts later, I had arrived at this:

5th Draft: I thought this would be the last draft, so I went a little overboard on the model making. The panels were glued together using a hot-glue gun, somewhat mimicking an epoxy fillet.

I built the model on a larger scale this time: I printed the pattern on paper and glued them on cardboard. The keel, frames and gunwales were cut from 1/8th inch plywood. At this scale, this was equivalent to 3/4-inch stock. I even made a wooden support and a double paddle. I broke 3 gunwales on that model before I succeeded to fit them. I only managed to fit them by starting from the middle, and then bending gradually toward each end. In the previous attempts, I had started at one end, bending toward the other end, with breakage occurring about 3 quarters through. I can't explain why the gunwales reacted that way, but I kept this trick in mind for the full-size build. I also decided to use half-inch thick material for the full-size gunwales.

I was happy with the look of that boat, at first. Then, after a while, a little detail bothered me. Although the top and middle plank were ending smoothly on the bow and stern, the bottom panel seemed to have a bit of a bulge in that area. I didn't know what to do next to correct that problem. I looked back at my drawings, and eventually an idea emerged. Looking at the end view, the chine angles used for the midship section were equal:

Chine angles A and B are equal for the midship section

What if the chine angles for each and every station were equal? So far, they were not. The difference in chine angle value just grew as you got near the bow or stern. The drafting process had been going a bit like this:

With a twisted V-bottom, if I used equal chine angles on all stations, the angle values would be different for each station. The drafting process would have to go a bit like this:

Doable but tedious. I didn't have a simple way of doing the task. I remembered a way to draw a boat using circles I saw on a web page here. I didn't know if that method could solve my problem. It was based on an Excel spreadsheet, a program I didn't have on my PC.

I played with circles and found a way to use them that work, but not 100%. The last couple stations still had to be done by trial and error. Here is a pictorial description of how I did it.

	Click to enlarge
Step 1- First, because the top plank is vertical (mostly), the sheerline and top chine are drawn in the plan view. The top segments for each station can then be drawn. The area between stations 0 and 2 is where the sheer, in the plan view, changes from an arc to a straight line. (Same thing between stations 14 and 16)

Click to enlarge

Step 2- Starting with the midship station (section 8), a circle (in blue) is drawn with 3 points touching the top chine, bottom chine and the lowest point on the keel. A copy of this circle is placed with its center on the lower chine point (dotted blue circle). A line (in blue), placed at the edge of this second circle, is drawn perpendicular to the bottom plank line. Its length is delimited by the first blue circle. Now, for a different hull shape, a different position for that blue line would have to be figured out. On the enlarged picture, red dots show the circle's 3 points.

	Click to enlarge
Step 3- Section 7. A 3-point circle is drawn (in green) with 2 points touching the extremities of the blue line previously drawn and the third point touching the top chine at this section.

	Click to enlarge
Step 4- Section 6. A circle is drawn (in orange) in the same way as the previous circle.

	Click to enlarge
Step 5- Section 5. A circle is drawn (in turquoise) in the same way as the previous circle.

	Click to enlarge
Step 6- Section 4. A circle is drawn (in yellow) in the same way as the previous circle.

	Click to enlarge
Step 7- Section 3. A circle is drawn (in gray) in the same way as the previous circle.

	Click to enlarge
Although circles can be drawn (in brown and dark green) for Section 1 and 2, the angle values achieved for the top and bottom chines are no longer close, probably due to the strong curve of the stems. Those last 2 sections have to be done by trial and error.

One beauty of this "trick" is that it automatically creates the curve for the keel. This trick is still a work in progress: the 2 intersection points for all circles were chosen by trial and error. For a different hull shape, such as less deadrise for the bottom, the position of those 2 points (the ends of the blue line in this example) would be different. A mathematician could probably perfect that trick, but I am not one. What matters is that method worked well enough to finish the last draft of my boat.

Did making the chine angles equal produce a fair looking boat? I think so. As a bonus, this new "rule" introduced just a little bit of hollow in the lower chine, near the bow and stern. Although the difference with the previous draft is subtle, it definitely corrected the bulge problem on the bottom plank. I now had a fine entry I was happy with.

Final Draft: despite my best efforts, the keel on that model was crooked, so I didn't go further than the minimum details. It was enough to judge the result. One year to arrive at this: OK, I'm not the fastest designer...

This is what the plywood sheet layout looks like:

Click to enlarge 1.- Paddle (half) 2.- Paddle center plate 3.- Breasthook

As you can see, there is not too much waste. In fact, even the waste area (in yellow) will be put to use. Splitting the paddle blades in half allowed to squeeze them in. The only additional wood required will be for the keel, inwales, outwales and seat. Plus the temporary frame and backbone used during construction. Below is an illustration of the construction, with the breasthooks removed to show more construction details.

Construction (Click to enlarge)

Here is a comparison between my design, that I called Baby Canoe due to its diminutive size, and a typical cruising canoe, according to W.P. Stephens.

While I was approaching the end of my drafting efforts, I finished reading an article that was bad news for my design. According to that article, a deadrise of 5 degrees is suitable for a stable boat, 10 degrees being on the sporty side. My design has 11 degrees. I knew that too much deadrise would mean an unstable boat, I just didn't know how much was too much. Now I knew. So I said to myself, ah well, it's gonna be a tippy boat. If I wanted to start building that year, it was too late to redesign. The hull shape of my design would be best suited for a keel boat perhaps, but my budget and space allowed only for a canoe, so a canoe it is.

According to an article in this magazine, my boat will be tippy, err, sporty.

That's it for now. Stay tuned for the next part, where we will begin construction.

On to Part 2

REFERENCES

SOFTWARE