Tandem Trike Problems
The major problems I encountered were: weight distribution; weight; chainlines; and gearing. Packing size was a bit of an issue, as was the complexity of the design. But mostly the problem was the sheer size of the beast. It really needed a car sized parking spot, or at least part of one (it could be hung over most cars in a garage).
The fundamental problem here is that the accessible load space is over the rear wheel. There is also the need to be able to ride the tandem by myself, which means that the front rider must sit over the front axles rather than in front of them. This means that with just two riders, the weight distribution is roughly 50/50 front to rear, in other words only 25% of the weight is on each front wheel.
With the load carrying ability being the rear rack, adding a load of touring gear just makes this problem worse. At times on CANC2 we had upwards of 50kg of gear on the rack, meaning nearly 200kg on the rear wheel. Luckily it was a good strong wheel!
Solution: I don't think there is one for a tandem tadpole trike. The options seem to be putting the stoker in front of the captain so that their weight is forward of the front axles, or putting load under the seats to move it forward a little. The problem with that is that the front seat has the stoker pedals under it, so using it for load space means making the trike even longer. Lifting the seat to fit stuff under it affects the stability and wind resistance.
The real solution is to build a back to back tandem, with the load between the seats. I spent a lot of time on CANC2 thinking about this, and decided it's the most elegant solution. It should fix the weight distribution and also make the trike lighter, stronger and have lower wind resistance. The only real problem is the increased drivetrain complexity, and the bike solution of separate drivetrains isn't really a good option on a trike I think (front wheel drive is hard to do). So my plan is to build a back to back bike, then think more about a trike. Ben Goodall at TriSled has built at least one B2B trike, and raced it very successfully. He wasn't happy about the drivetrain though, and has expressed reluctance to build another one.
My tandem weighed about 50kg, according to Ian Humphries who actually weighed it. That's a lot, and given more experience with building stuff (ie, after building this and a couple of other things) I could easily shave 10kg off that weight. The extremely oversize main tube was a bit excessive, and in retrospect I should have imported a length of the 65x1.4 4130 that I found online.
Solution: Really, the whole design is a bit silly if you want to minimise weight. Either go to a twin tube design like the Organic Engines Troika (thanks Dan), or accept the flex that comes from a light single tube like Greenspeed do. Of course, my choice is to build a different trike altogether, but that doesn't really solve the problems with this layout. I don't know of anyone with a B2B tandem trike in regular production, which may suggest something about the practicality of the idea.
I had the usual first project problems with chain lines, most notably fitting the drive side chain in around the front seat tubes. The chain also dragged on the ground in some gaers, but htis could be fixed by running it over the idler at the rear, so I did that most of the time.
Solution: With a couple of tweaks and an extra pulley I managed it, but it was never very tidy in either single or tandem modes. The Greenspeed tandem approach is to run the stoker chain forward to the captain, then a huge long chain back to the rear wheel. This works, and avoids my problem of moving the bottom brackets around when swapping modes. But neither solution is really very good in my opinion.
The huge range of gears needed for a tandem recumbent tourer is going to be a problem whatever you do. Spped range from below walking (1kph) to over 100kph (downhill into Perth), although yfou can probably ignore pedalling above 70kph without inconveniencing the riders too much. I bought an 80T chainring but never actually fitted it, as the 60T off my racing trike proved sufficient. Which is good, because those big chainrings are fragile.
Solution: The Mountain Drive (website is unusable without Flash plugin, sorry) worked really well, and being able to shift it while stationary saved people a lot of trouble when they stopped in the wrong gear. Aside from that, a megarange cluster and 42T/60T cluster worked well.
The thing was not trivial to design, build, ride or maintain. Assembly really required a power screwdriver to deal with the 30-odd allen bolts that hold it together, and tuning everything so it actually worked was basically impossible. I never got the rear gears to index reliably.
Solution: Ignore the problem. Seriously. With 4m from the gear lever to the rear derailleur any clicking will be hidden by the panniers and the distance. I ended up mounting the front derailleur lever on the derailleur post, so that only the stoker could shift it since it proved impossible to do it reliably from the front seat anyway.
The trike was about 5m long, and 1.2m wide. It weighed 50kg. Even in bits, it was too big to fit under a bed or in a box. It was BIG. I couldn't store it anywhere, and it spent time hanging in a tree as well as parked in other silly places.
Solution: Dismantle it and use the parts in other projects. That's what I've actually done, because it's too heavy to ride regularly, and too big for me to store permanently. A lockable car garage in Sydney costs $50-$100 per week, and long-term storage places still charge about $30/wk for something that big. For someone who has a car space already it would be less of an issue, as it would be possible to hang the trike above most cars in a standard space. Richard Guy Briggs has cycle storage at his apartment block, which is great too.
The plan was to have a trike that could pack up nice and small into bike boxes or similar and be shipped around that way. Not so - the center section was too long (2.2m), and the trike was a bit on the heavy side. But that wasn't really an issue, as I never flew anywhere with it. And since I built it, Greenspeed have come up with a 5 coupling version of their tandem that does everything I set out to do, at about the cost of mine. So it can definitely be done.