(click the plans to see a larger version)

A recumbent is a bicycle where the rider sits in a recumbent or semi-recumbent position (technically, recumbent means laying down, which would make it a little difficult to see where you’re going). So, usually the riders legs are in front of them. Think of sitting on a couch. This relaxed position lessens the strain on the neck, shoulders, arms, lower back… and so on. The major downside is that it’s hard to lift yourself off the seat for balance, which is necessary for mountain biking and BMX.

Back in 2004, it had been my intention for some time to build a long wheel base recumbent. If you do not know what a recumbent bicycle is or want more information on them, one of the better sites to start at is the IHPVA site. Check out the “builder’s corner” part of the site. Recumbents (or ‘bents) tend to be expensive due to the low volume of sales, so there are many homemade bikes out there.

Chromemoly can be welded with oxyacetylene. I did not have much luck with oxyacetylene fusion welding or brazing of aluminum. I expended a good deal of effort teaching myself to TIG weld aluminum, only to decide against using it. Still, I’m glad I know how to weld aluminum now, even if the skill does not have immediate applications.

My friend Ben drew the plans of the frame of the bike out (see, mechanical engineers do have a useful purpose in life) after I told him my concept for the bike. I liked the idea of one main tube running the length of the bike. The rear wheel would be attached with a suspension to make calculations easier (there is a great deal of stress on the rear end of the bike– having a pivot, spring, and shock absorber makes life easier since you can just beef up the spring if its too soft). The front wheel would be controlled by an overseat steering wheel attached to a universal joint at the steering tube. The rear wheel would provide the drive, with an idler gear under the seat to keep the long chain in place. I looked at front wheel drive bikes but they are too complicated for a first bike. The problem with such bikes is that having the drive on the same wheel as is used for steering can make the bike difficult to steer unless it is built properly.

I wanted to try adding an electric motor to the bike. There are several ways to do this: a friction drive, belt drive, or chain drive. The first two I threw out immediately because of their inneficiency. Since the bike cannot carry many batteries, it must be very efficient. The chain drive must be attached to drive the bike somehow. There are some difficulties with this. If the motor is simply attached to the same wheel as the pedals are attached to via the chain, when the motor engages it will spin the pedals. A freewheel must be added to the pedals to prevent this. The converse situation- that of the pedals spinning the motor- can be a problem, depending on the motor. I bought a motorcycle starter motor. The drag it causes is negligible. Another solution to the drive problem is to have front wheel drive for the electric motor. This is simpler than a pedal powered front wheel drive system, as there is no crank and pedals to take into account. The motor can be mounted on the fork above the wheel. A rear wheel would be used for the front wheel since it already has gears attached. Once again though, FWD may interfere with steering. Another option is to have an intermediate drive where both the motor’s and the pedal’s chains meet before driving one of the wheels. The best method will likely be determined by experimentation.

With an electric motor, batteries must be taken into consideration. They are bulky and heavy and there must be adequate space for them and thought in the design to keep the weight distribution roughly even and the center of gravity low. With our recumbent, the main tube is located high enough off the ground so that two batteries can hang underneath the tube and the rider’s legs. This lowers the center of gravity and keeps the weight distribution close to even.

The electric motor will require a controller. A starter motor cannot be run at full power continously. A microcontroller will control a MOSFET and pulse the motor, changing the duty cycle to change the power delivered to the wheels.

We finally built the recumbent in July of 2004. Low carbon steel was used. The only steel we bought was the main tube, which was 2×1″ square tubing, 1/8″ wall. It is entirely overkill. 1/16″ would suffice and save weight. I insisted on using only one donor bike and this resulted in quite a few compromises. However, I believe this is a good thing as otherwise the complexity of the bike would have gone out of control and it would be less practical to build more of these bikes. We did away with the suspension and welded the rear triagle from a Magna bike to the main tube. Next, we attached the head tube with a 38 deg angle. We added a mount for the steering: a metal tube inside schedule 40 PVC. A bolt was run through the inner tube and cables with tensioners run to the steering tube. Next, we added the pedals and used 2.5 normal bike chains from Walmart. The new chains are hard to splice because the pins holding the links do not come out easily except at the ends so we threw in half of a chain from an old Schwinn in the middle. We thought about making our own light weight seat but decided to sacrifice our welding chair (a computer desk chair) instead. Very comfortable! We rolled our creation out onto the street and realized immediately that it was unrideable for reasons similar to the chopper. There was too much oversteer. Back into the shop it went and a 5inch offset was added. This lowered the oversteer a bit, but neither of us could ride the bike further than about 20 feet. Back into the shop. We thought about it and decided that the head tube angle was far too great. We changed it to 18 degrees, which is the head tube angle on Ben’s Schwinn. We rolled it out onto the street and realized that the cable steering would not work now because of the nearly 90 degree angle between the steering axis and head tube. The cables broke when we tried to ride it anyway. We decided to go with the simplest solution and welded the handlebars from an exercise bicycle on (another timely dumpster find). This tiller steering is not ideal but it is completely solid and very simple. Just place the handle bars just above your knees and weld a tube from the handle bars to the post coming out of the head tube. One note… keep a spray bottle of water handy as the heat from the torch will ruin the bearings located all around the bike! Have someone keep these cool with the water while you weld nearby.

We could not find an idler gear so we cut a channel in a block of Delron (Teflon substitute) and bolted the block to the boom just beneath the seat. We also swapped the 26″ front wheel for an 18″ wheel. This unfortunately made the front brakes unuseable. A post was welded onto the top of the boom between the riders legs where the gear shifters were attached like the verticle shifters on road bikes. To this day I have routing troubles with the cables to the shifters however.

We slapped on several coats of Krylon and the bike was ready to go. A front derailleur and bar in the rear for luggage was added later. The bill for the bike came out to $25 for the boom and other metal (half of which was used for other projects) and another $25 for the chain, brake cables, and krylon. We bought far more than we needed anticipating that we would build another bike.

A final frame will be built out of chromemoly. As a note, it is cheaper and less of a hassle to buy an $80 full suspension bike from Toys’R'Us than it is to round up all the parts separately. The quality of the parts is mediocre, but the point is that cost should not keep you from building such a bike.

This bike is a joy to ride and gets as much attention from bystanders as the chopper. Braking is poor (since we never mounted a front brake) and the gear shifter has cable routing problems. Otherwise, the tiller steering is the only hurdle. I much prefer the recumbent style to an upright bicycle, and hardly ride uprights anymore, unless my route requires off road travel or has many hills. My one other gripe is that the gearing doesn’t go high enough. Unlike an upright, where pushing hard lifts you off the seat, the recumbent riding position lets you harness all that work. It’s easy to max out the gearing.

Test rider Shane easily mastered tiller steering and went for a ride.

Closer view of the whole bike. Note how, after half a year, rust is attacking even the vestigial galvanized steel pipe above the crank. This pipe is actually handy for lifting the bike into my truck. The front wheel is actually on backwards. The gears should be on the opposite side if the electric motor were attached.

Closeup of the rear triangle. I attempted several brake upgrades before settling on Shimano V brakes from a donor bike. Brake hardware has many little incompatabilities. Lack of attention to braking is the greatest deficiency of the bike. Especially with the added speed boost provided by the combination yellow and electron blue paint job.

The Delron idler. The chain is not in the groove as it should be in this picture. The delron would wear too quickly to use this bike for a daily commute.

The crank, showing off the massive chain. The weight of the chain, and the drive system as a whole, is a weakness of long wheel base recumbents. Bicycle shops like to tell you that you will need expensive, lightweight cranks, gears, and chains made from exotic materials. We would like to say that, if you mount an office chair to your frame, you can probably ignore a few extra ounces in your drive system.

The shifter positioning is reminscent of road bikes. Except that road bikes dont have a pointy ending. Hence the duct tape. A few minutes with a power tool would fix that problem. Cable routing is the greatest nuisance on this bike.

Closeup of the steering. Note the scars from several surgeries.