So the basic idea was to build a carbon fiber reinforced, hollow plywood core recumbent, based loosely on the design of Woodie. Next I had to put the new design on paper. I downloaded a free CAD program called TurboCad and worked out the design. Here's what I came up with:
Note that the seat is not a separate piece but actually an extension of the frame. Without a movable seat (or boom) there would be no way to adjust the reach to the pedals. Obviously, this was a risk. However, I had moved the seat around quite a bit on Woodie so I was pretty certain I knew exactly where I wanted it. On the plus side, I would never have to worry about the seat sliding out of position! Also, I figured this would save some weight by eliminating the hardware for attaching the seat to the frame.
I also eliminated the rear suspension. This saved about a pound and also reduced the overall length by allowing the rear wheel to be tucked in closer to the seat back. It balanced out the weight distribution too, which had been biased to the front on Woodie.
I had become very comfortable with Woodie's steering, so I kept the same head tube angle and trail for the new design. I was comfortable with Woodie's 28 degree seat angle, so I kept that the same too.
One design objective was to minimize the vertical deflection of the chain at the idler. Riding Woodie I had noticed that the chain drive felt the smoothest and most quiet when using the small cog-small chainring combination. I attributed this to the fact that the vertical chain deflection was least in this combination. Also, minimizing the deflection of the chain would minimize the load on the idler axle, reducing the possibility of bending or breaking it, or worse tearing it out of the frame.
At the same time, I wanted the seat to be as low as possible to minimize aerodynamic drag. I had originally hoped to have the seat at about 16 inches (quasi-lowracer territory), but 19 inches was as low as I could go without inducing more chain deflection than I wanted.
One mistake I made on Woodie was in the orientation of the front derailleur tube. I did not angle it back far enough. As a result, when I used the small ring the chain would drag on the bottom of the front derailleur cage. It was actually fine with my original 30 tooth ring, but when I tried to go to a 28, it started to drag. I corrected this in the new design by orienting the front derailleur tube at an angle of 69 degrees relative to the chainline, as it is on a typical road bike.
Also shown in the image above are the areas of the frame I planned to hollow out. Those areas would ultimately be reinforced with carbon fiber.
This drawing shows my plan for the cross section of the hollowed out regions. Compared to Woodie, I reduced both the height and width of the beam section by about 1/4 inch. Also, I reduced the sidewall thickness from 1/4" to 1/8" and the top and bottom wall thickness from 3/8" to 5/16". Of course, I expected that the addition of carbon fiber would more than compensate for these changes.
Jump to Woodwork