Building A Canoe: The Finishing and Floating

At the end of April it was finally warm enough to pull out the canoe and finish it. All that was needed was some final sanding and varnishing. I went over the inside and outside with my Festool sander using 120 and 180 grit pads then hand sanded with 150 and 220 grit before brushing the first coat of varnish on. I choose System Three spar varnish because it is what they had at WoodCraft and I didn't want to buy it on-line, pricey stuff though, I paid $38/quart. I choose the satin finish because I thought the canoe didn't need the extra glitz and because satin doesn't show scratches at much.

System Three Spar Vanish

I guessed that two quarts would be enough without doing any calculations and, luckily, I was right (Gil's book didn't provide any guidance). I was able to apply three coats, inside and outside, plus four coats on three paddles and used only 1.5 quarts. The label says 350 to 450 sq-ft/gallon but I seem to have gotten more than that (assume an average of 3 ft wide x 16 ft x 2 sides x 3 coats = 288 plus three coats on the seats, yoke, and paddles equals about 300 sq-ft for .375 gallons = 800 sq-ft/gallon). I did thin the stuff with paint thinner a couple of times but mostly because of evaporation during the application. In any case, two quarts seems to be the right amount for most canoes.

I used a 3" Purdy synthetic bristle paint brush and didn't notice any brush marks -- it dried smoothly. I didn't sand between the first two coats because I put them on within 12 hours but then sanded the with 220 and 320 grit sand paper (by hand) before the final coat.

Here is the finished piece. The final weight is 66.6 lbs (I lost 0.2 lbs in the final clean up, mostly in the decks as I had to get them to match the level of the shell). Given that Gil said this design would weigh in at approximately 65 lbs, I'm pretty happy since I used lumber that is much heavier than the cedar he uses. I'm sure I saved much of the weight in the seats.

side view

front view

Bow Deck

Stern Deck

Bow Seat

Stern Seat

Seat Spacers

Yoke

Stern

Over the winter I prepared three paddles, (58", 63", 68") out of my mountain of old barn flooring that I have stacked in my backyard. I wasn't sure it was a good choice but it was free and at the very least it was good practice material. It turned out to be a good choice  in terms of looks, weight (I got them all down to 2.6 lbs), and performance, but dealing with all the knots made it difficult to shape and probably doubled my working time. I wasn't sure how to make them but soon realized that it is just the same as any sculpture, remove material with finer and finer tools until you have the shape you want. The key was in copying the plans from an old paddle that I liked and using the shape, transferred to a board, to rough cut it the shape on a bandsaw. Then using a draw knife followed by a spoke shave to get an approximate shape. Unfortunately, the high number of knots in the lumber made those tools particularly difficult so I eventually started using a grinder to work through the knots and get the blade as thin as I needed (about 1/4" at the edge). Then my Festool sander working 40 grit to 180 along with a rasp to get the handle shape. The final step being hand sanding. 

The feature I'm most proud of is the thumb notch on the handle. I haven't seen it on other paddles but I found that it really make them much more comfortable to hold.

Note the thumb notches!

You can see that the blades on the right and left had nails, which were rusted by years of horse pee, that cause a black stain -- I really like the look. I filled the holes and knot gaps with black epoxy. 

Horse pee/rusty nail stained blades

I suspect all the lumber is old growth spruce but there is no doubt with the middle paddle. I tried to count the rings but had a hard time. My estimate is that, just on the handle, there are approximately 115 growth rings (plus or minus 5) within a 3" radius. I'll need to find someone with better eyes to get an exact count.

Approx. 115 growth rings  

So everything was ready for a test voyage on May 4. I took it out to Green River Reservoir, about 30 miles north of me. To be honest, I was a little nervous about whether it would float and how it would perform but it turned out to be great in every way. I'm very happy, not only in how it looks, but how it performs as well. One thing I learned pretty quickly though, is that when using it as a solo canoe, the bow and stern flip (since it is symmetrical this isn't a problem) because I need to sit closer to the center to keep the bow from tipping up too much. The paddle also worked great (I used the 68" one). I don't think I have ever used a paddle that was actually the right size for me, so it was a real revelation. I also found the large blade to provide much more power than I've ever found in a paddle before. I highly recommend making one's own paddles, in comparison to what you can buy, these are much better.   

As a final note, my wild life count for the first voyage was several mallard ducks, two loons (my first -- these are super cool birds to see and hear), a couple of kingfishers, and one beaver. Not bad for a three hour cruise.

Green River Reservoir, May 4, 2013

Building A Canoe: Steps XI, XII, XIII, XIV

Part XI: Make and Install Gunwales

After finishing the fiberglass work, and cleaning up the annoying, unsightly, ridges, it was time to install the gunwales (or rails). Because the longest board of ash I could find in the lumber yard was only 12 ft, I had assumed I would have to splice two sections together. However, before I had implemented this plan, I learned of a friend of  a friend who had a 20 ft board of ash available. Luckily, I was able to use this person's shop to mill the lumber as well because there was no way I could have done it in my shop. In fact, even with a 12 ft board I was going to have to go to a friend's shop. It would have been helpful, however, if I knew the final length of the gunwales ahead of time because of the cumbersome nature of milling such long strips. I thought the curve in the canoe would result in strips longer than 16' but, it turned out, my outwales were only 15' 7" in and my inwales were 13' 7". I also milled them differently than Gil suggested. He suggested 3/4" by 3/4" for the outwales and 3/8" x 3/4" for the inwales; but I didn't see how 3/8" was wide enough to install the seat hardware, which are 1/4" carriage bolts, so I made mine 1/2" wide. With the outwales, I thought 3/4 stuck out too far so I trimmed it down to 5/8".

My first big mistake, though, was in trimming the outwales. I wanted to cut them close to the actual length because I was taking them to my shop to pre-drill and countersink the screw holes used to attach them. In lining up the outwales, I marked it from the top of the stem, and when I started to install them I realized that the canoe gets longer as you move down the stem, so it left me short to get the entire outwale covering the stem. Luckily, I had added a little to the length to make up for any small error, but it wasn't enough. I hoped that by pulling the canoe together, making it more narrow, I'd be able to squeeze another 1/2" out of my outwales, which worked well enough that in shaping the ends, it looks decent enough and maybe even mistake free. 

Pulling the canoe tighter to make up for the short, pre-cut, outwales

Close-up of the short outwales -- Oops! Next time remember to mark from the bottom!

After the shaping, they don't look too bad, and maybe even intentional

Unfortunately, I realized when I returned my canoe forms, and looked at the lender's canoes, that I didn't need to attach the outwale with screws from the outside, that the screws from the inside were enough. In my defense, in looking back at Gil's instructions, he doesn't explicitly say this and the images show the outwale being attached with screws at the stems in order to also hold the deck in place. The one hint he gives is that he says you need 65 to 70 screws, placed every 6 inches, to attach the gunwales to a 16 ft canoe and I thought his math was just bad when I realized I needed around 115. I could have just left the outwale screws in place but, in addition to the aesthetic issue, I thought it created an increase risk of rot at each of the hole openings so I filled them with walnut (and the occasional cherry) plugs. Problem solved and I think it ends up adding to the aesthetics rather than detracting. By the way, Gil recommends brass screws but my experience with them is that they are just too soft and the heads break off all the time (even with pre-drilling) so I used stainless steel. I had absolutely no problem with broken heads and I think they look just as good.

Mistaken screw holes filled with walnut and cherry plugs

Part XII: Make and Install Yoke

For the yoke, I used a figured piece of yellow birch. I made the shoulders on it especially large, which I think works well. I had never carried a canoe on my shoulders before and the first time I lifted this one I had it resting on my neck. Big mistake. That doesn't feel too good. Better it should rest on one's shoulders, especially with a life jacket on.

Curly yellow birch yoke

Part XIII: Make and Install Decks

For the decks, I created a couple of blanks out of quartersawn ash and walnut veneer. The walnut is between each strip of ash, and as I cut and reglued the sections, I added additional pieces of veneer to create a complicated fault system (I suspect this aesthetic results from the lingering effects of many undergraduate geology classes).

Deck blanks of ash and walnut veneer

Gil recommended created a cardboard template for marking out the shape of the flush deck. I highly recommend this method, it worked out well.

Flush deck installed (bow)

Flush deck (stern)

I didn't know how big the decks would be, so I made blanks much larger than I needed and turned the scraps in to art.

 Three Triangles

shellac and acrylic paint on ash and walnut

16.5" x 17.5" x 1"

2012

Three Triangles (closeup)

Part XIV: Make and Install Seats

Of all the aspects of my canoe, I'm most proud of the seat design (provided that it turns out to be comfortable for many hours of continuous use, which I won't know for several months). It is light weight, strong, easy to make, original, and, I think, very beautiful.

I didn't want to make a standard caned canoe seat, partly because I didn't want to be responsible for the furniture joinery and partly because I couldn't see spending 20 hours weaving it. In order to avoid that monotony, I mulled a few different ideas around for several months, including a Wendell Castle inspired sculpted seat but realized that that would be too bulky. Somehow the idea of making a bent lamination seat popped in my head, which seems perfect in strength and weight. Having never made any bent lamination work though, I needed to experiment a little to get what I wanted (of course, I could have looked on-line, but for some reason, I didn't bother, which is just as well, too many people recommend using urea-formaldehyde glue which is just too toxic for me to want to work with. Instead, I used my standard Titebond III which really is the worlds best all-around wood glue. It worked fine). For the stern seat, I glued 4 thin strips together (each about 3/32" thick) in the form. For the bow seat, because the gap is greater, I used five strips.

Bent lamination in form

(this is an after-the-fact staged photo, hence, no mess or plastic keeping the form from sticking)

I wondered how many rows would be comfortable and thought that I would need three, but when I tested two, I didn't think that an additional one would have any advantage so I'm sticking with two until I'm proven wrong. I used some butternut that I bought from my neighbor when he cut down a tree in 2009. I've had it air drying in my shop ever since, which makes it perfect for a project like this because it is less brittle than kiln-dried lumber that I would have gotten at a lumber mill (maybe it doesn't make a difference but I like to think it does). They seemed to have plenty of flex in them, making them more comfortable, and they seem plenty strong too. I've stood on the front bow strip (160 lbs) without any negative consequences, so I figure two of them should be able to handle the weight of anyone getting in my canoe.

Stern seat (bent lamination butternut)

For the spacers, I played around with a scrap piece of maple and more walnut veneer. I think it is an interesting detail and it really didn't take that much time to make.

Seat spacers (maple and walnut)

The final activity I worked on before stopping for the winter was to clean up the top edge. I had thought that a spoke shave would be the best tool for the job but it turned out that my 102 Lie-Nielsen low-angle block plane worked best (gotta love that little thing!). Now I just need to wait for spring to do some finish sanding (on the decks), touch up epoxy work (on the decks, gunwales, and seats), and then add a few coats of varnish. Until then, it is going to rest in the middle of my living room (which is fine, I don't like using my front door anyway). As is, it weights in at 66.8 lbs. Not bad considering Gil says this canoe should weight about 65 lbs and I used hardwoods that are much more dense than white or red cedar.

An almost finished canoe

Building A Canoe: Steps VIII, IX, X

Part VIII - Fiberglass The Outside
After cleaning up the outside, the next step is to seal the wood with epoxy before starting the fiberglassing. I'm using West System's 105 resin and 206 slow hardener. For this size project, Gil Gilpatrick says you would need two containers of the "Group B" units, which is about a gallon container of the 105 resin. I really like West System's measuring pumps, which makes accurate dispensing so easy. The directions say that slow hardener shouldn't be used if temperatures go below 60 degrees (the directions warn about it possibly not curing at all), so it was fine for September, but I switched to the fast hardener, 205, for fiberglassing the inside, which I did in October. Actually, I found that 206 still works with night temperatures going down into the forties, it just takes longer. Anyway, the sealing process is the easiest part of the whole project, it just took a few minutes to mix up the epoxy and spread in out in a thin film over the hull.

After lightly sanding the surface, I reinforced the stems, as Gil recommends, with two layers of fiberglass on the bottom part of the stem and one layer on the front section. This is good practice for learning how to work with epoxying fiberglass. My work wasn't perfect but it can be sanded smooth enough and another layer or two of fiberglass goes over it anyway.

 Reinforced Stem

I should have taken a picture of the fiberglass before epoxying the hull. There are actually two layers on the bottom, one being an (American) football shape, the other covering the entire outside. I found the process of wetting out the fiberglass with the epoxy and getting it smooth to be very stressful because of the importance of finishing before it starts to harden. It is probably best done with two people, preferably with at least one person knowing what they are doing, but, all the same, it was doable. The biggest problem I had was in rapping the fiberglass around the stem. I ended up with a bit of a mess, partly because of inexperience but also because, when I read Gil's instructions I didn't turn the page and look at his picture, which would have clarified exactly what to do. Oh well, I figured I could clean it up with my Festool sander when it hardened, which I did. Problem solved.

It ended up taking about 3.5 hours (with a temperature of 77 degrees at 1:00 when I started) to wet-out the fiberglass (with epoxy squeegee cards) and by the time I was done, the first part had set up too much to be manipulated any more. I should have gone over the first part with a roller when I was halfway done, so the job wasn't as smooth as I would have liked.  

 Fiberglass Outside

After cleaning up the purely amerature, but passable, epoxy job, spreading out a second coat of epoxy was easy. Gil says that a third coat may be needed to fill all the fiberglass fibers, and mine were definitely not all filled, but other people I've talked with recommending using as little epoxy as possible because it adds too much weight and just looks bad the more there is. Better to clean it up and add many coats of varnish after the second layer of epoxy, so that's what I'm doing. 

 After Two Coats of Epoxy

Part IX - Clean Up The Inside
In turning the canoe over, I finally had to deal with cleaning up the inside -- not a particularly appealing task given the difficulties I had created for myself.

Start Cleaning Inside

Here is a better view of the hundreds of dowels (approximately 1500) I had sticking through the hull. I trimmed them with my dozuki (Japanese pull saw) before pulling out my Festool sander/grinder again. Because it was impossible to trim them flush, it was difficult to keep the remaining dowels from tearing up the sanding pads. Eventually, I managed to get it reasonably smooth with 40 grit paper before switching to 60 grit. After 60 grit, I started filling gaps with epoxy again, and then cleaned that up with 80 grit paper. I went over it with 120 paper before going back and rubbing it out with 100 grit by hand.

The hardest part was cleaning out the stems. I had to use chisels and coarse paper to get it reasonably clean. As long as no one is inspecting it with a flashlight, it should be good.

Unfortunately, the canoe flattened out as it rested on the saw horses without the forms. I added some support but it really didn't help much.

Part X - Fiberglass The Inside
So I put it on the ground before sealing with a first coat of epoxy. Again, pretty darn easy process. I used one of those cheap disposable Chinese brushes to get epoxy in the stems.

 Inside Sealed with Epoxy

I had to sign it before epoxying the inside and debated how to date it. I'm not sure that I'll finish this year but didn't want to pre-date it either. I decided to be optimistic and sign it 2012.

I decided I'd better pull the sides together before fiberglassing the inside for fear it would be too difficult to do after the fact. I used the forms to be sure I got the dimensions right.

I was fairly nervous about fiberglassing the inside. Gil's description made it seem pretty difficult and he ends the discussion with what I preceived to be hallow words of encouragement -- "You can do it!" I took this to mean, "I can't do it and it will look shitty" and thought it might make sense to do it in two sections that overlapped. My neighbor, who hadn't built a canoe but has plenty of fiberglass experience thought this was a good idea, but luckily I ran into a friend, who has built a canoe, the day I was going to do it and he talked me out of it -- just one piece, one time. 

It went smoother than I expected and only took 2 hours. The inside stems where much easier than the outside because Gil recommended just cutting the fiberglass short and not worrying about getting it in there.

I thought it looked pretty good when I was done and I had time to go over it several times before it set up but, unfortunately, the next morning I noticed a lot of ridges in the fiberglass. I guess I didn't pull the fabric tight enough as I was wetting it out. Oh well, again, this is what the Festool sander is for (and no, they aren't paying me to say this but I certainly wouldn't mind if they did). Of course, sanding it smooth really thinned out, or totally removed, the fiberglass in some small areas, but I'm thinking that on the inside it isn't going to make much of a difference.

Unsightly Fiberglass Ridges  

And here it is after the second coat of epoxy. It has settled a bit again. Though the width should be a little less than 35 inches, and it is about 38 here, I should be able to pull it back into shape when I add the yoke and seats. 

Fiberglassing Done!

Building A Canoe: Steps V, VI, VII

We'll, the cool weather has finally started to creep into Vermont and now I feel an definite urgency in working on the canoe if I want to get it into the water before it ices. Once it gets below 50 degrees, I'll have to bring it indoors to finish the epoxy work but I'd really rather not have it set up in my living room. 

Step V - Finish Stripping
I found the stripping process to be very slow. It would take between 1.5 to 2 hrs to complete a full row all the way around. No doubt it would have been faster had I just used one type of wood, or even if I had a set pattern with different woods, but, counter-intuitively, there is much more thinking involved in trying to do something that looks random; that, along with getting small sections to fit tightly, changing between drill (definitely need to pre-drill the holes for screwing down the strips) and driver, and dealing with all the clamps, it was just a very slow process.  

I finished stripping on August 18. The last row was a little tricky because I needed to shape two pieces with a bead on both sides. They also had to be sculpted over a length of about ten inches from 3/4" wide to a point but I found my Lie-Nielsen 102 block plane was the perfect tool for the job. It was great to see them slide into place perfectly. The final count was 69 rows. I had planned on using only one strip of cherry but ended up using three, one on each side and one on the bottom. In addition, I had been worried about not having enough strips but, in the end, I had plenty, with 350 feet left over. In retrospect, 1200 linear feet would have been perfect (I started with 1385) but I'm sure I can find a use for my surplus. I'm thinking of making a flat wall piece with them. I think it would look great to have them displayed next to each other. 

Step VI - Fill Holes With Dowels
After stripping, my next step was to drill out all the staple and screw holes and fill them with 1/8" and 3/16" maple dowels. I had no idea how many 36" dowels to buy so I was glad to have a local source. I ended up making five trips to Spooner Specialties, picking up a total of thirteen 1/8" dowels and nine 3/16". I didn't count my holes but if I assume the my dowel length was 1/2" per hole, I filled over 900 1/8" holes and over 600 3/16" holes -- probably pretty close to the actual but I'm not interested in finding out the exact number. By the way, fair warning to any copycat canoe builders out there, forcing the 1/8" dowel into the 9/64" hole I drilled into the basswood was not fun. It was a real tight fit because the basswood didn't cut cleanly (I think the humidity of summer had something to do with it but filling the walnut holes was so much easier). My blisters have at least healed but I still have a sore shoulder.

I needed to take the canoe off the forms early in order to finish putting the dowels in the stems. 

In addition, I didn't feel comfortable with the design of the stems. I felt they needed some additional reinforcement. Ideally, there would have been a laminated strip inside the stem that the strips could be attached to, but it was too late for that so I ended up sticking 5/16" dowels through both sides and epoxy-ing them in place. I think this will work as well as anything. 

Step VII - Cleanup and Fill Gaps
The next step was to clean it up, shape the stems, and fill any holes/gaps. One of the problems with using walnut is that it didn't flex as well as the basswood so I did have a fair amount separation on the tight curves. Nothing too extreme, but definitely noticeable. I tried to make an epoxy putty that would match the wood using PC-11 (marine epoxy) and Transtint's Dark Mission Brown dye. For some reason it came out more gray than I would have liked but it was good enough.

It took about three rounds of sanding and refilling the holes. Luckily, I have a real nice Festool sander/grinder (it can be partially seen on the right side of the last image). Between cleaning up all the dowels, glue, and epoxy, I don't think I would have wanted to use anything else. I started with 36 grit paper (touching very lightly), and worked down to 120 grit. Gil Gilpatrick said that anything smaller than 80 grit doesn't make a difference but the sander/grinder leaves a very distinct sanding pattern so I wanted to go smaller. In addition, I went back and hand sanded with 100 grit paper just to make sure it looked good.  

 And there it is, prepped and ready for the first coat of epoxy!