2005 Frame Design and Construction


The frame design process began with focusing on our main design parameters. We wanted a light and easily maneuverable vehicle with a very small frontal area. We decided on a recumbent style bicycle due to the success of these in many of the IHPVA's races. One of our main influences was Warren Beauchamp's Barracuda.
We began working in Autodesk Inventor with a human solid model. Since the rider was the one thing that we couldn't change we designed the frame around the physical limits of the rider. To keep the frontal area as small as possible we opted for a medium wheel base and the rider's legs straddling a 20" front wheel.
We wanted to keep the frame as light as possible so we designed the frame with no triangulation to minimize the amount of tubing. We used the finite element package PATRAN/NASTRAN to optimize the tubing diameter needed to keep the stress in the frame at an acceptable level. The final choice was a 56mm diameter, 1.1mm wall thickness, 4130 Cromoly steel tube which we obtained from Burley Design Cooperative.
We used weld design methods to analyze the critical weld that joined the chain stays to the main frame. It turned out that the stress in the welds would be unacceptable, so we redesigned the joint to remove the majority of the stress from the welds.
Once we were satisfied with the frame analysis the model was made as accurate as possible. The fork, bottom bracket shell, dropouts, and chain stays were donated or reduced in price by Gordon Freedman at Cycle Classics in Portsmouth. Gordon builds custom bicycles and gave us a lot of valuable advice on the frame construction. The donated parts were added to the 3D model. The only items that were not entirely accurate were the wheels. We had to finalize the model before the new wheels came in, so we used the dimensions of some similar spare wheels we had. This ended up causing us problems when the new wheels showed up.
After everyone was satisfied with the model we made a set of working drawings to use at the shop. We plotted the drawings on 'C' size drafting paper for easy visibility when we were working.
Eric, the team machinist, began by coping the ends of the tubing to accept the rear assembly and the bottom bracket shell. He used a boring bar setup due to the large diameter cutouts. We wanted a fairly nice fit up since we had decided on TIG welding the frame, so we took our time to make the cuts fairly accurate.
After Eric coped the tubes, Jason set up a small fixture on the chop saw to hold the tubes at the proper angle while mitering the ends.
Kendrick and Eric are working on the rear assembly. We inserted the chain stays through the rear tube to take the stress out of the welds. The holes had to be drilled at an unusual angle so the chain stays lined up correctly with the rear dropouts. The holes were eventually oversized some to allow the chain stays to stay parallel. This left a some gaps that were a little tough to TIG weld.
This is the first mock up of the frame. All of the tubes have been machined and are ready for welding.
Kevin Colvin at the ODU Model Shop did most of the frame welding for us. He started with the rear assembly. We built a wooden jig to hold the assembly in place while welding. Wood may not have been the best material, but it ended up working without burning too much.
This is a shot of the rear assembly weldment. The chain stays were inserted through the tube and fillet welded on each side. The side that mated to the rest of the frame had to be ground down some to make the fit work.
We used standard vise grip clamps and some scrap material to make a fixture for the frame. The welding table was a flat 3/4" steel plate and we used that as the reference plane. Some stops were welded to the table to hold the tubes at the correct angles during welding.
After we welded the front assembly, we needed to make a bore for the head tube. We mounted the frame to the back of the mill table with about eight different clamps to keep it secure while we bored the hole. An extra long boring rod was needed to get all the way through the tube and we had to keep an eye on all of the clamps because they vibrated loose easily.
This shows the head tube inserted into the new bore. It was little tough to cut the bore but it ended up working and made a nice precise fit. The head tube was custom made for the Ritchey head set that we purchased at a reduced price from Mike Shipp at East Coast Bicycles. We couldn't find a head tube of suitable length and material from a frame dealer, so we made our own. We didn't bore out the entire tube to a thin wall thickness and it ended up being a little heavier than necessary.
The frame was smoking like a dragon while the head tube was welded in.
This the fixture we used to weld the rear weldment to the front weldment. Angle blocks were used to the keep the frame perpendicular to the table and shims were used to keep the chain stays in line with the frame tube.
A similar fixture was used to weld the underside of the frame.
This is the final frame weldment with the Dahon aluminum fork inserted into the head tube. It looks just like the model, except for a couple of errors. We realized that we welded the bottom bracket shell in backwards and that some of the tubes were cut approximately 1/2" too long. These errors didn't cause any headaches, because the bottom bracket can be used backwards, and the tube length didn't change the frame dimensions too much.
Derrell is putting our first coat of paint on the frame. Kendrick and Jason both wanted a yellow frame, but they ended up compromising and made it partially black. This finish was just to keep it from rusting while we worked on it. The real paint job would come before the competition after we had finished working on the bike.
After the second seat was designed a rear triangle was added to the frame. This provided a mounting area for the rear brake and the seat. It also helped improve the strength in the rear section of the frame.
The frame is hanging up at Flatlander's Hot Rods, where Aaron Bloch sandblasted the frame after all the welding had been completed. We painted it with some cheap silver spray paint that didn't quite give the expected finish, but that's what happens when you paint it two days before the competition.

Back to Vehicles