The Bike Canoe Trailer

​by Christian

At some point, early in the pandemic, Line lamented that we didn't have a canoe trailer for our bikes  - vehicle access was closed to most beaches, but bicycle access was open. If only we could tow the canoe there on our bikes, she was thinking... Little did she know (well, actually, she probably knew) I've been wanting to make a bike-towable canoe trailer for quite some time. This isn't really going to be a step-by-step with plans, as I mostly winged the design and measurements were written on the back of a cereal box I've since lost... but if you're crafty it will probably give you some ideas.

Materials

This is a heavy duty trailer - all welded-steel construction. I made it out of some scrap from work, left over when our operation moved to a new building. The transverse bars the canoe rests on are 1" sched 40 pipe, and the longitudinal bars are 1" square HSS with 0.150" wall thickness. Rather than sticking dropouts for the wheels right on the bottom of the longitudinal bars I made short "forks" (from the same 1" HSS). This give the canoe a bit more clearance overall, and (importantly) angles the tail up in the air a bit providing a much better exit angle. It was also pretty easy to "carve" dropouts in the bottom of the forks with a drill and angle grinder making for easy wheel mounting. The tow bar is lightweight by comparison - it's 3/4" pipe with 0.070" wall thickness. But the tow bar is only ever in tension/compression - the canoe itself provides the structure in that direction.

The trailer weighs 65 lbs, so it's not light (but neither is the canoe - it's a 20' Clipper MacKenzie, and weighs 98 lbs. Plus I often throw another 100 lbs of paddling accessories and camping gear in there. Add F and N on the back of the bike itself for another 80 lbs or so...). I used some 26" wheels from a "department store grade" mountain bike an old housemate left behind (thanks Sonja!).

How does it tow?

Well, it's certainly heavy. I definitely notice it going up hills. But it's remarkably stable, and I can roll it (carefully!) off or onto a curb. Assuming you load the canoe such that the centre of mass is in front of the wheels (ie - there is "weight" on the tongue and would tip forward if not attached to the bicycle) it is very stable - even coasting downhill at 40 km/hr.  Note that the weight needing to be forward of the axel for stability is a property of trailers in general, not just this one in particular. This is a cute little demo video (note that it's by people who sell a tongue-weight measurement device... they give a reasonably narrow range of acceptable tongue weights. I'm pretty sure that trailers are still stable at heavier tongue weights, but there are other reasons not to put the weight too far forward - mostly the tongue weight your vehicle, or in my case bicycle, can handle.). I just wing it when loading, but make sure that the centre of mass is far enough forwards that it's still ahead of the axel even on the steepest of hills. The trailer also has automatic breaking, and a mechanism to keep it from rolling backwards when stopped facing uphill (more on those later).

Automatic Braking

I realized early on, given the weights involved, that I needed some kind of brakes to not get run over by my own trailer. The trailer and canoe were going to weigh as much as me, before I even put any gear in the canoe.
The length of the canoe made cable-actuated brakes seem like a bad idea - I'd need at least 3m (half the length of the canoe) plus another 2m (the bike) for a total of 5m of cable. The longest cables one can easily find are for tandems, but they are only 3.5m long. Of course, one would need a splitter at some point anyway, to run one brake per side, so you could imagine doing that in the middle... but cable-stretch would surely be a problem too. One could imagine hydraulics, perhaps even running metallic automotive brake lines for much of the distance, but then I need to mess around mounting/unmounting the lever every time I unhook the trailer. A friend of mine went the hydraulic route for his canoe (and raft) trailers.
I started wondering if I could avoid these hassles all together, and eventually decided that I should just make the tow bar itself actuate the brakes. It would be guided on the trailer, but not rigidly attached. When going forwards I'd make a little catch that pushes on the frame, and in the other direction it would pull on the brake cables. I thought I was a genius until I discovered that this concept already exists, at least for rental automotive trailers (it's called a "surge brake").
At first I tried to set it up so the tow bar was constrained with bushings, but it's hard to get the alignment perfect when welding and I had problems with the system binding. So I just welded up some guide loops with a bit of slop (out of the same 3/4" pipe that I used for the tow bar) on the underside of the trailer and that problem was solved. I doubled up the tow bar under the trailer and made the guide rectangular so it can't spin around. I have another contact point on the "bow cradle", and that one got a bushing (made from melted down milk jugs). It's a bit loud rattling around in there, going over bumps, but the slop doesn't actually matter when it comes to towing as the single bushing close to the bike (where it matters most) and the length of the canoe makes it track straight.
For the first iteration I decided that I wanted to give the brakes as much mechanical advantage as possible, so I didn't feel the weight of the trailer when slowing down. To accomplish this I put the cable-stops way out at the edges of the trailer, so I had a huge trigonometric term acting in my "favour". This worked way too "well" as I discovered quite quickly on my test rides around the block. When coasting it was possible that the canoe might want to coast a little bit faster than me... then the trailer would slide forwards on the tow bar (as intended) until it reached the brake cables - a gentle nudge on the tow bar would exert a huge tug on the brake cables, locking up the wheels on the trailer for and slowing it down in a hurry. Of course I was still going forwards, so the brakes would almost immediately disengage and I'd quickly use up the travel on the tow bar before smashing the tow-catch into the frame, lurching the trailer forwards again. It was possible that the trailer would "bounce" forwards from the impact and then hit the brake cables again setting up increasingly violent oscillations. 
For the second iteration I reduced the mechanical advantage substantially, basically putting the cable-stops right beside the tow-bar guides. It's still possible to get an application of the brakes when coasting, but it's mild and doesn't set up violent oscillations. I definitely don't feel the full weight of the trailer when slowing down and when stopping hard downhill I've found the tires of the trailer lock up before before the tires of the bike, so I think it's a good balance. I am considering adding a bungee or spring to hold the trailer (lightly) against the tow-catch when I'm not braking hard, but haven't bothered to do so yet.

I made brake mounts from leftovers of the 1" HSS by screwing in some M10x1 brake posts through it into some M10x1 nuts tack-welded inside (since they were hard to reach with a wrench). I made a jig for mounting them straight with a block of wood. The cable stops are just replacement barrel adjusters for brake levers; since they aren't under very much force (and are actually pulled onto the locknut anyway by the cable tension) I just drilled and tapped holes for them on some relatively thin scrap pieces of steel I welded on beside the tow-bar guide. I got the angle right by mounting the barrel adjusters, aiming them towards the brake catch on the tow bar, and welding in-situ.
I thought about making a little bushing for the brake catch on the tow bar, such that it could rotate and "balance" the braking force between the left and right wheels... but it actually turned out not to be too hard to adjust the brakes to catch at the same time. I just gently push the trailer backwards by the tow bar, such that only one side catches, and then tighten the cable on the other side a little using the barrel adjusters. Eventually I can get it so that both sides brake evenly. Thus far it seems like wear keeps things even (maybe this makes sense - if one side is braking harder those pads should wear down faster.

Anti-Rollback

Another fear, due to the weight, was that I'd never be able to get going if I had to stop on a steep hill. Or, worse, that I'd loose control and get dragged backwards downhill. Since I was pillaging the wheels off of a bicycle one of them is a rear wheel, which comes with a handy built-in freewheel mechanism, making the solution to this problem pretty easy. I just cut the largest cog off of the cassette and welded it onto the frame a little ways ahead of the wheel, like a stationary chainring (spaced out a bit by another piece of scrap, to make room for the chain). I cut an appropriate length of chain off the donor bicycle, and joined the bottom with a bungee cord. Only the top is in tension when preventing the trailer from rolling backwards, and the bungee gives me enough slack to easily flick it on/off the cogset (so I can disengage it and roll the trailer backwards when I need to, for instance when parking). Works great! It does have a bit of a tendency to want to turn when the anti-rollback stops it from rolling downhill, since only one side is stopped, but the bicycle easily counters this if pointed anywhere close to forwards (as it would need to drag the bike sideways). I can even hop off the bike entirely, put the kickstand down, and leave the bike unattended facing uphill.

Hitch / Bow-Cradle

The tow bicycle had a Chariot hitch on it already (most of our bikes do), so I wanted to use that. At first I hoped to just buy the "lollipop" direct from Thule - this would make the job super-easy - but it's no longer available as a replacement part. A very unfortunate result of the sale of Chariot to Thule was that Thule decided to discontinue replacement parts. When based out of Calgary you could get literally anything on there as a replacement part for (almost) free (this let me make my own ski attachments using replacement pop-in axel from some dumpster skis and aluminium scrap for under $20). The smallest sub-assembly I could buy that contained the lollipop was the entire bicycle tow-arm for something like $100. This would basically double my budget, so I decided to make my own. (I did consider just making a socket and using the entire Chariot tow bar, but we often use the Chariot and the canoe trailer on the same trips!) My first iteration was just a rope with a knot in it, but it was possible for the knot to pull through the chariot knuckle under load. So I made a little piece of aluminium that fit snugly into the chariot knuckle and drilled a smaller hole through that for the rope (exiting on the diagonal, to leave room for the lock-pin down the back-centre that traps it in the knuckle). I poked the rope into the tow-bar, fired a bunch of screws through it, and voila! I found that the rope was getting chewed up, so slipped a piece of plastic tubing over it at some point. Seems to be holding up well with the tubing. The extra slop from the rope isn't great, but it's not too bad, and it provides some flexibility. I am considering trying to injection mold my own lollipop, by making a mold from the lollipop from our Chariot and plastic from an old pair of ski boots.  I guess I'll post an update if I ever do that.

The bow cradle was a second-stage iteration that prevents the tow bar from slipping off to one side or another (or bending) if the canoe tips forwards (the first iteration just had a 3" piece of sched 40 pipe ratchet-strapped to the bottom of the canoe under the bow as a guide). It's a longish piece of sched 40 pipe, with some saddles hammered into the shape of the prow and a pice of angle welded on at the correct angle. A ratchet-strap goes around the front hand-thwart and down/across to webbing-guides on the cradle (just little beads of weld). It's very stable - you can pick the canoe up by it. It comes almost (but not quite) right up to the hitch when the canoe is mounted at the correct position on the trailer - enough travel for the tow-bar-actuated brakes to do their thing, but leaving very little unsupported length of the tow-bar poking out the front. At the very front tip is an HDPE bushing (that I made from old melted-down milk jugs) so the tow bar it still slides smoothly even when the weight of the canoe is bearing down on it.

DIY Tips

Some tips:

• I guess this is a general welding tip - but make little jigs for everything, so that things assemble straight, and clamp well. I made a lot of jigs out of wood, which is fine since I was only making one trailer and it didn't matter if they got a bit burned. Blocks to get the spacing/alignment right for the longitudinal wheel bars/dropouts were key (and actually having the wheels in place when tacking).
• Tack first, then check assembly, then tack all sides, then check assembly again, before welding the crap out of it.
• A coat of paint makes it look way better, and hides all your welding sins.
• I made the dropouts by drilling holes of the correct size and at the right spot, and then cutting a slot over to them with the angle grinder. It's way easier to get things in the right place with a drill than a zip-disk.

Regrets / Future Modifications

• I put the anti-rollback device on the wrong side - this was totally an oversight and a result of winging the project, rather than doing a design and making plans. It would be way easier to activate/deactivate if the wheels were switched, such that the chain was on the outside of the trailer. It's not enough of a pain to bother changing anything, though.
• I'll probably make a "deflector" or some kind of angled protrusion on the front corners of the trailer. I discovered (fortunately during low speed maneuvers) that it can really catch up on edges if you get too close and clip a bollard or post (and if the anti-rollback is engaged this can really get you stuck). I think that clipping something at speed would be catastrophic because the trailer would just stop dead. If the part of trailer that contacted the post was angled, rather than 90 degrees, it might just shimmy off to the side instead. This would have been easiest/cleanest to accomplish by just extending the outermost HSS members a bit and bending them over towards the middle during initial construction, but I can always weld something else on later.
• The trailer is really wide (mostly because the canoe is really wide). It barely makes it between the bollards guarding most bike paths. I sometimes wonder if I would have been better off making the trailer narrower and installing it further back on the canoe, where the canoe is narrower (and with a longer tow bar). Then the canoe itself could have been the widest part. It also would have let me put the trailer itself inside the canoe if I ever wanted to paddle it around. Would have let me feel less comfortable putting tons of gear in the canoe, though, since it would be hard not to have most of the weight riding on the hitch.