Vancouver, B.C., Canada
June 2020: Gathering tools and supplies, and lots of thinking
While watching some of the early prep videos for the course I’ve been plotting my way through a tools and supplies list and settled on some things I need on my shopping list. One great Ebay find was a $40 Stanley block plane from the 1930s that just took a little cleaning and tuning up and immediately impressed me.
I’ve been trying to source bending oak locally here. I found a great supplier on Vancouver Island but couldn’t make sense of a dedicated ferry trip to go get it. I played around with other local options and (re)discovered yellow cedar. It is often used on the West Coast in small boat building and is traditionally used by the local First Nations here in fine carving and steam bending applications. I dug through some Engineering studies and did a lot of math. I have settled on some rib dimensions that I think will compare to the stiffness of Brian’s Oregon white oak. It’s all theoretical so far and definitely straying from the course a little, but I would like to give it a shot.
4th July 2020: Buying the wood and some pre-prep work
Having settled on yellow cedar I’ve now found it almost impossible to buy unless I special order a log to be sawn up for a small fortune. After lots of phone calls to suppliers it turns out that stumpage rates have gone up, a mill closed for pandemic-related reasons, and the main supplier here shut down a couple of years ago. I finally found some in 16-20’ lengths of 5/4 (“five quarter” which varies but in this case is 1”x5.5”) air dried, totally clear, straight and rift grained and excitedly picked up a few boards.
20th July 2020: more prep work and learning to steam bend
Saw horses, zero clearance inserts, and a steam box! Having just built those I’ve now been able to do two rounds of steam bending tests using Brian’s benchmarking method. After a few exploding pieces, I’m now getting really good results with fresh cut, then soaked yellow cedar steamed for 2-3 minutes. Only small tear outs when really pushing the bends like the first rib on the canoe. I can’t tie yellow cedar in a knot…at least not the thickness I need to use, and the bending oak is clearly superior from what I can see compared to my wood, but the yellow cedar is working. There are two nagging things have been concerning me about this project – steam bending and ripping long lengths, particularly the curved gunwale laminations. So far making these tests has really got me feeling excited about the steam bending. Brian’s videos are incredibly helpful.
August 8th: Ripping!
I set up a 16’ infeed and 16’ outfeed for my table saw. I learned a couple of things…getting all surfaces totally level is very helpful to avoid the end of a long piece wandering out and disturbing the cut; thin kerf blades are magical. I switched blades as an experiment. So far I can see no downside to using a thin kerf 7-1/4” blade on the table saw other than depth of cut. It flexes just a little which if the workpiece swings out as I mention above keeps the cut from wandering, saves wood, makes less noise, cuts more efficiently and costs less.
I should mention that this project is clearly taking a long time…I’m slowly working at it on purpose as I’m keeping it a surprise for my wife…hard to do with the setup of some of these tasks!!
August 16th: Thickness planning by hand…
I don’t own a thickness planer… I would love one but in absence I’ve been using a hand power planer. Its working incredibly well with sharp blades, going slow and checking thickness with a caliper. I also built a plywood jig around it to try to make a thickness planer jig. It doesn’t work well but it helps to get a better feel for staying square on the faces and a bonus is its going to help trim the ribs to an exact ¼” at the ends later. So its stayed on…
August 22nd: Laminations!
Having watched the updated videos for this process over and over I followed all the careful tips on sighting the length along the worksurface, flipping clamps around to get the curve to stay true (I actually weighted a couple of the bar clamps to get more leverage). The glue up went perfectly and had just the right amount of glue thanks to the discussion on the videos. I cut my shear curve blocks exact instead of adding for the centre standoff (!) and had to rush over and grab some 2×4 offcuts to get the right height mid-glue up. I took the clamps off at 3.5 hours at ~28˚C and the chisel worked really well on the mostly-hard glue.
With the help of a neighbour supporting the work, we managed to rip the gunwales perfectly. I used only one 1”x5.5” board for all the material so had almost no room for extra. The thin kerf blade was again perfect here. I found that standing WAY back to aim the lamination at the blade to start feeding, about the middle of the curve, then walking it forwards, feather board in place the whole time, it worked really well. As I previously mentioned this step was a big concern to me and I’m glad to say it worked out well.
September 6th: Morticing the gunwales
Again, thanks to some excellent ideas and instructions from Brian on the videos this went really well. I bought a used plunge router off Craigslist and a spiral upcut bit from Lee Valley Tools (really good price by the way). My only challenges were with the old router. First was my fault, I didn’t check the plunge mechanism when I bought it. It had totally seized. I overhauled the whole thing and had to ream out a nylon slider bushing for one of the shafts. Once that was done and I started into routing I noticed the bit was creeping deeper into the cuts! I ended up at about ¾” deep on one before I noticed. It turns out the chuck, being old, wasn’t quite grabbing the ½”-¼” reducer and the bit was slowly able to move a little. I put a huge wrench on the chuck and that solved that…
September 12th: Stems
With this being a larger boat, I knew the red cedar 1x8s for the stems would not work. I didn’t have access to yellow cedar in 1×8 size. I debated other species but I wanted something very rot resistant, and so drawing on the good experience from the gunwale glue up I decided to laminate two of the standard 1×8 stems back to back, from the same board cut mirror image of each other. The thought here is that they would work against each other to resist any cupping. I threw on 40 clamps because…why not! After I get the stringers placed, I’m going to thin and shape these for a finer water entry, keeping the inside parts wider, probably the thickness of the keel. I’ll still taper the keel and the base of the stems as Brian suggests.
I also got most of the last bits of prep out of the way – rounding stringers, building spreaders and capture forms, cutting the gunwales to length. I won’t rip the ribs up until I’m ready for them as I want them to be fresh sawn.
September 26th 2020: Hand carving a paddle
This really has nothing to do with the canoe build, but I decided to hand carve a canoe paddle for my wife. I was going for a deepwater paddle, so no risk of dinging it on rocks, and wanted to use western red cedar. I found a really beautiful 2×6 with a huge knot in what turned into the centre of the blade, and decided to go for a modified otter tail, with a nod to Brian’s techniques and shaping his Greenland kayak paddles. The knot and changing grain patterns made it incredibly difficult to carve but I persisted carefully and I like how it turned out. It’s obviously more fragile than a cherry paddle… No idea how it paddles yet, but it looks and feels really nice!
October 3rd 2020: Starting to look (a little) like a canoe!
It’s getting exciting now! I built the spreaders, separated and spread the gunwales apart, kerfed and tied the gunwales ends and got to check the symmetry of the gunwales and deck with a string line. Its really fun to see a curved piece of wood I’ve been carefully measuring and routing turn before my eyes into the outline of a real canoe with some great looking proportions!
I really found the kerfing more difficult than I think it needed to be. After I was almost done I realised that the capture forms were too tight on the gunwales and relaxing them to where the gunwales JUST touch then sawing made things so much easier.
For fun I really hastily clamped on the keel (flipped on its side so the clamps could hold it) and the stems just to get a quick look before packing up. the rocker measurements were also just roughed in for a photo.
Late October/Early November 2020: Milling ribs and cutting them to length
Recently I’ve not found much time to work on the canoe, so this phase has been a series of 20-minute tasks strung over many days. After some experimenting and consulting with Brian I’ve settled on 1.41 as the r/b ratio. On paper it seems a little fuller than I was originally going for (1.40), but test fitting up some mock ribs it looks like the right way to go. For reference…I feel that a freighter would be around 1.44 and a classic Prospector would be around 1.42 with these dimensions.
I started with a single 16′ length of yellow cedar (in which I found a single knot!) and then ripped and planed it down to 36 pieces of 1″ x 23/64″ x 54″. After that I found the rib rounding jig incredibly useful dressing the pieces.
I used a 3′ long piece of wood for my rib measuring stick (the canoe is 36″ wide). This allowed me to scribe a trace of the canoe width onto the stick at each rib location, on one side of the stick, and mark out rib adjustments on the other side. This made for really fast work with no room for error in measuring the ribs to final length. I decided to cut ribs to length on the table saw to get clean 90˚ cuts. My home-brew planer attachment worked out really nicely to get every rib to exactly 1/4″ at the ends. Looking back I think a similar result could be achieved just by clamping the planer base onto a piece of 1/4″ masonite (which I did also) but I’m telling myself this extra effort was a little safer. If you read this and want to try the same just treat it like a table saw – I guess it has the potential to shoot ribs out at speed, although I didn’t have any of that happen.
After some test bends in the summer I’ve found that soaking for at least 3 days before steaming gave more control on the bends, so after building a stegosaurus in the front yard to test fitment of every rib mortice, I have laid all the ribs down under water (and a little light snow) for a few days.
November 19th: Rib day!
West coast outdoor workspace = tarp city! Hey, I was running out of patience waiting for a day without rain, of which there are very few here at this time of the year. So, I got all set up in the back yard, rain started after a few ribs right on time, and I ended up getting the last rib into the mortice as a thunderstorm erupted. It made for a dramatic end to the day but the boat stayed dry.
I was really happy with how the steam bending went. I was particularly nervous about this whole plan not working at all at this stage. I had taken a bit of a gamble trying to get this done with yellow cedar rather than the suggested Oregon white oak. Things went very well. If someone else was to choose a non-standard wood I really recommend just taking your time, doing some testing and paying attention to the wood thickness based on its strength, to get something at the end of the build that is comparable to the oak in overall stability.
So as mentioned the ribs are yellow cedar, like most of the rest of the boat, and it took a little R&D to figure out steaming parameters. Dimensions I settled on were 21/64” thick and 1” wide. Based on the aforementioned “way too much math” part above. I soaked the ribs outside (rainwater top-up optional) for two weeks until I couldn’t wait any more. I settled on 2:45 steam time for the first lot and once the steam box was really hot, around 1h30 in, I cut that down to about 2:30 a rib. This was a REALLY fun day watching the boat take shape. For the tighter bends I backed the ribs with a length of seatbelt material. I found in my tests that a leather belt I had wasn’t wide enough to be able to secure the edges of the rib so well and the grain would tear out. Seatbelt material has almost no stretch so that helped.
I ended up cracking one rib out of 33 which I thought was really good. I had only given myself three spares. Over the next couple of weeks I ended up changing out four more for a total of five replacements, two were for shaping issues, and the end ribs came out of the formula too short. I ended up adding 3/4” to the end rib replacements (3/8” either side) and making them out of 1/4” stock to get a tighter bend than the other rib thickness would allow. By the time I had bent in the last ribs the spares had been soaking for a month. I trimmed the spares to length right before steaming and noticed that the water from soaking had only penetrated into the wood about 1/16”! maybe the cedar won’t even need to be oiled?!
December/January: tweaking the rocker
No longer building in secret! I’ve I did a bit of a reveal to my wife a day after steaming which was fun. Its pretty hard to hide a full-sized canoe frame covered in clamps for more than a day or two! Things have slowed down a lot with this build, just because of the weather really, and only having space outside to work. That’s fine I’m not in a rush.
I’ve been playing with progressive rocker, and settled on bending in a keel depth at the outside part of the stems. This really helped with the lines I think. I’ve also been trying to figure out how to actually measure rocker. Turns out to be a bit philosophical and no clear answer! In an attempt to figure out if 3 inches is three inches is 3”, I’ve found that the best sets of plans that accurately break out rocker for various canoe styles are on The Carrying Place website. They have a nice adaptation of an original prospector on there based on survey of an early model, amongst other things. I’ve settled after some looking at baselining rocker at ~90% of waterline length which seems to be the most often used (explanation here). With that, and progressive rocker I’m now at 3” 90% WL with an aim to land at around 2-1/2” when the stringers are on. When I started into this I designed in a “planned” rocker of 3” projected out to full length (untrimmed keel 16’4” long). That same measure was now 4-3/8″ with the progressive rocker.
January/February: lashing on the stringers…twice!
Still slow with the weather, don’t take this slow progress as this being a difficult build, its not. I lashed on all the stringers at the end of January and then to my surprise found that I’d bent out almost all the rocker. I was left with maybe an inch at 90% WL, really not what I was going for, probably still a “canoe” but not the plan. After some discussion with Brian and different options for how to proceed I decided to pull off the lashings (350 in all…) and re-do the stringers again, pre-curving the shear more to compensate.
To avoid the same thing, I put the deck on horses at even thirds (before they were at 25% from ends to match the un-stressed shear curve, my mistake), and weighted down the ends of the gunwales with ~2lbs each end. This had the effect of over-curving the shear from the un-stressed 7-9/16” to an even 9”, the theory being the tension of the stringers was pulling the shear (and therefore rocker) flatter. For anyone else’s reference later…I measured the shear curve changes while lashing. Pretty consistently I lost 5/16” of shear (and rocker) for every pair of yellow cedar stringers I lashed on.
At the end of the second lashing (that’s 700 lashes now! Before and after photos below) and the keel clamped at every rib I have a good-looking rocker curve of 2-3/8”. I expect with the keel on (later) this will go down to ~2-1/4” at 90% WL. My shear curve is at 7-9/16 which is exactly where it was before lashing. Cool.
I also looked at the lines for symmetry and took a few measurements. I’m really happy with the left-right symmetry of the whole package. There’s some minor left-right asymmetry at the one end where some of the hull shape might have a slightly slacker bilge on one side to the other, that ends up with the stringers not exactly the same distance from the keel. That’s my theory anyway. I’m done messing with the shape now…maybe.
I was really surprised how efficiently the lashing went. I’m also surprised at the rigidity in the frame with all the lashings on. The design of Brian’s overall building system is seriously cool.