Reports of the demise of this Web site are greatly exaggerated! We at thank Harris Cyclery for its support over the years. Harris Cyclery has closed, but we keep going. Keep visiting the site for new and updated articles, and news about possible new affilations.

M.I.T. Aluminum Bicycle Project 1974
find us on FB
by Harriet Fell and Marc Rosenbaum

This article appeared in the Proceedings of the 27th International Cycling History Conference, North Haven, Connecticut, USA, 2016

Spoke Divider

In 1974, Professor Shawn Buckley ran an M.I.T. Independent Activities Period course where participants, including Marc Rosenbaum and Harriet Fell, got to build their own bicycle frames. Marc Rosenbaum, then a student at the Massachusetts Institute of Technology, set out to create a bicycle that would give a rider an acceleration advantage by being ultra light without sacrificing stiffness. In this paper, we describe some of Marc’s design, the frames he and Harriet built in the IAP course, and how these played a role in the Klein vs. Cannondale patent litigation in the mid-1980s.

Marc Rosenbaum’s Aluminum Bicycle

In 1974, Marc Rosenbaum was a senior at M.I.T. majoring in mechanical engineering. The goal of his undergraduate thesis (Rosenbaum 1974) project was to create and describe a bicycle that would give a rider an acceleration advantage by being ultra light without sacrificing stiffness. Marc stressed stiffness over strength, pointing out that “frames never break in normal racing conditions, which indicates that strength is not a problem.” Clearly things have changed since carbon fiber bicycles have entered the racing scene though the problem is not their resistance to normal loads encountered during rides but to impact damage.

In 1974, normal track bikes weighed 18 or 19 pounds. There are lighter bikes today; the I.C.U. minimum weight limit is 6.8 kg, which is about 15 pounds. There were also some lighter bikes in 1974. A bicycle made for Eddy Merckx in 1971, which he rode to set the one-hour record (30.715 mi), weighed only 13.25 pounds. It had a very light steel frame, custom built titanium handlebars, stem, and seatpost, partially plastic pedals, 24-spoke wheels, and tires expected to last at most 4 hours on a smooth board track. The components were extensively drilled out. This bicycle was not designed for strength or to last (Rosenbaum 1974).

Marc's Bike
The bicycle built by Marc Rosenbaum weighing 12 lbs. 5oz.

Marc’s Design Guidelines

Spoke Divider

Spoke Divider

The Frame

The table below shows the tube lengths and thicknesses that Marc used in his frame. The frame was put into a jig to hold it in alignment and was joined by tungsten-inert-gas welding. After welding, it was heat-treated (annealed) to relieve stresses induced by welding and to bring the aluminum back up to its initial strength. The top tube sagged about 1/32 of an inch during annealing. This was apparently due to the greatly decreased strength of the aluminum at the high temperature of the heat treatment process.

Frame - Tube Specifications
TubeOutside diameter (in)Thickness (in)
Top tube1.50.049/.032
Head tube1.50.049
Down tube1.50.049/.036
Seat tube1.50.049/.036
Bottom bracket1.750.187
Seat stays0.6250.625
Chain stays0.7500.083 taper to .045

The Bottom Bracket

The bottom bracket was bored out to press fit the bearings directly into it. The result is a highly accurate and simple system. A Teledyne titanium axle was modified to accept the precision bearings. The axle saved three ounces and the bearings were 2.5 ounces lighter than the conventional bearing arrangement.

The Hubs

The body of the rear hub is of 2024-T4 aluminum. It was bored out for precision sealed bearings and has a hollow titanium axle. It weighs only 5 ounces vs. 11.5 ounces for standard track rear hubs. Similar construction was used for the front hub. It weighed only 3 ounces vs. 9 ounces for a conventional hub.

Marc's rear hub
Marc's rear hub
Marc's front hub
Marc's front hub

Spoke Divider

Spoke Divider

The Pedals

Marc's pedal
Marc's pedal

The pedals were designed to be comfortable, easy to get into (with toe clips), and to allow sharp cornering. They have platforms of 2024-T4 aluminum on which the whole ball of the rider’s foot can rest. The axle is of titanium and is threaded to fit a standard crank. The bearings are pressed into the pedal body and onto the axle. The pair of pedals weighs 7.5 ounces, 5 ounces less than the lightest track pedals of the time.

The Saddle

Marc's saddle
Marc's saddle

The saddle is a prime example of eliminating unneeded adjustability. Marc determined his saddle-to-pedal distance on another bicycle and decided on an integral seatpost/saddle for the bike. The result is a savings of nearly a pound.

Handlebars and Stem

Marc's handlebars and stem
Marc's handlebars and stem

Track bikes have the bars down low to put the rider into an efficient sprint position. Track riders often use only one position. Marc attached the handlebars directly to the fork crown, reducing weight and increasing rigidity. A simple, wide shallow U-shaped handlebar was attached via the handlebar-mounting end of a steel stem silver-soldered to a steel strip that was in turn silver-soldered to the top of the fork crown. This arrangement saved almost a pound.

Other Components

The wheels used Hi-E Engineering tubular rims and Pirelli Specialissimo Corsa Leggero tires. The track sprocket, chain, and T.A. cotterless 48 tooth chainring were standard. The bike initially had an 18-tooth rear sprocket for a gear of 72 inches. The headset was Stronglight Competition, and a Raleigh road-racing fork was modified to give the required wheel clearance and the design fork rake of 1.5 inches.

The bicycle that Marc built for his undergraduate thesis project was the world'’'s lightest track bicycle at that time. The finished bicycle was as rigid as a normal sprint track bicycle and with wheels built up for road riding, weighed 12 pounds, 5 ounces. The bicycle is currently at the M.I.T. Museum.

Spoke Divider

Spoke Divider

Harriet Fell's Aluminum Bicycle

In 1974, Harriet Fell was an Assistant Professor of mathematics at Northeastern University, across the river from M.I.T. The (M.I.T. Independent Activities Period) is a special four-week term at M.I.T. that runs during the month of January.2 In 1974, Professor Shawn Buckley offered an Aluminum Bike Project class. This was an opportunity for students to build their own aluminum bicycle frames. The tubes suggested for the class were of a smaller diameter and had thicker walls than the ones Marc decided to use. Harriet showed up and ordered a set of tubes.

She designed her frame by using some measurements (e.g. top tube, seat tube, and the angle between them) from her favorite bicycle, a Holdsworth Strada. Working with the constraints of the wider tubes and her small size (5' 2") and making sure to leave enough space for 700C wheels, she made an accurate half-size drawing of the planned frame. She then measured the lengths of the other tubes from the drawing. With advice and instruction from Marc, Shawn, and the machine shop staff, she learned to use a metal lathe and milling machine enough to machine the tubes and dropouts for her bicycle. She put the parts into a jig, and Shawn sent it off for welding. Shawn said that the completed frame was particularly light and strong. (The fact that it was made for someone only 5' 2" tall surely had something to do with this. A frame for Eddy Merckx at 6' 1" would surely have weighed more.) Shawn demonstrated the strength of the frame by laying it on its side and standing/bouncing on the seat stays.3

Gary Klein also made a frame in this project in the spring of 1974 and got even wider tubes to make a larger frame.

Harriet’s Frame Goes to France

Just after finishing the frame, Harriet moved to France to continue her career in mathematics. She also started cycling with the cyclosport group at the Faculté de Science at Orsay. She got as far as mounting a headset and fork on her frame, but math and cycling, including riding the 1975 Paris-Brest-Paris, filled most of her time. In September 1976, unable to extend her leave of absence for another year, she returned to Northeastern University. As she was determined to move back to France, she left her frame, a large bag of books and clothes, and a pile of tubular tires in a friend’s cellar. She did spend another year in France, but not until 1988/1989 with her husband, Sheldon Brown, and their children Tova and George. The clothes and books were intact. The tubulars had hardened to a rock-like state and were unusable. The frame met another fate.


In 1983, Cannondale manufactured its first aluminum bicycle (Wikipedia 2016). Gary Klein had been issued patents on aluminum bicycle frames and sued Cannondale for patent infringement. Schwinn was already making large-diameter aluminum frames and paid Klein license fees. Cannondale claimed that Klein was not entitled to his patent(s) because the real design came out of Marc Rosenbaum's thesis and the 1974 Aluminum Bike Project (Bike Forums mtnbke 2009)(Justia 1989) Marc Rosenbaum was subpoenaed in the Cannondale/Klein case and gave a 6-hour videotaped deposition. He hid his bicycle in a friend’s barn so it wouldn’t be subpoenaed. He was afraid he would lose the bicycle for years and possibly never get it back if it were taken.

Defendant Exhibit sticker on Fell Bike
Harriet's frame
Defendant exhibit in Cannondale case

Cannondale’s lawyer wanted frame(s) that were built before Klein’s for the case. He contacted Shawn Buckley who told him that there were five light/strong frames produced in the project. The lawyer called Harriet to ask if he could use her frame and she said, “Yes, but it’s in a friend’s cellar in France.” He visited her friends and brought the frame back for the trial.

Bringing in large diameter aluminum frames built by Bill Shook, founder and engineer of American Classic , and by Harriet Fell that predated Mr. Klein’s, Cannondale successfully argued that the Klein patent was null and void due to "prior art" - you can’t patent something that’s already in existence and known publicly (Bike Forums tcs 2009).

Klein's patent discovery documents had referenced Marc's bike and thesis numerous times, and it had taken Klein almost eight years to be granted a patent. Essentially, Cannondale asked, "Why is Klein's bike different from Rosenbaum's frame?"

From Frame to Bike

Marc was right to fear losing his bicycle for many years if it were subpoenaed. Cannondale kept Harriet's frame for many years and never let her know where it actually was. Eventually she asked her husband, Sheldon Brown, who worked in the bicycle business, to call a friend at Cannondale and ask for the frame. That worked. When the frame arrived, Sheldon hung it high on a living room wall and told every visitor about it.

In May 2005, Sheldon asked Harriet what she wanted for Mother's Day. She said, "Sheldon, since we've been married you've built up about 40 bikes for yourself (Sheldon Brown's Personal Bicycle Page) so how about getting that one on the road for me." He built the aluminum frame up with standard, fairly light-weight parts. Harriet has recently changed the saddle to a Brooks Cambium and added lights, a Quad-Lock mount for an iPhone, and a small frame bag with USB charger, all easily removable.

Harriet's Bottom Bracket and Crank
Harriet's bottom bracket and crank
Harriet's Dropout
Harriet's dropout
Harriet's Brooks Cambium saddle
Harriet's Brooks Cambium saddle
Harriet's bike with iPhone, bag, lights
Harriet's bike with new additions

Spoke Divider

Spoke Divider

Proof in the Riding

Marc Rosenbaum rode his bicycle up Mount Washington twice4 before donating it to the MIT Museum. Bill Shook rode his homemade bicycle frame in several races (Justia 1989). Harriet Fell continues to ride hers around the western suburbs of Boston.

Marc on his bike in 1974
Marc on his bike in 1974
Harriet on her bike in 2005
Harriet on her bike in 2005


1. From correspondence from Marc Rosenbaum: "For solid tubes, section modulus is proportional to radius (or diameter) cubed, so strength increases with the cube of diameter. Moment of inertia is proportional to diameter to the fourth power, so stiffness increases with the fourth power of diameter.
For hollow tubes, it's not as cut and dried. If you double the diameter and double the wall thickness, then strength goes up as the cube and stiffness as the fourth power. If the wall thickness remains constant, doubling the diameter doesn't increase strength by 8 and stiffness by 16."

2. It is an opportunity for MIT students, faculty and staff to organize or participate in a wide variety of courses ranging from Blacksmithing to Charm School.

3. Harriet’s frame did not show any sagging due to annealing. The tubes on her frame were shorter than those on Marc’s bicycle and had slightly thicker walls.

4. He did not ride this with the original 72" gear. He rode Mt Washington in 1975 alone as the race organizers wouldn't allow him in the race with a fixed gear bike. He rode a 1:1 gear - 20T rear and made the front chaining from a Sturmey Archer 3 speed cog (so about a 27” gear). The next year he rode the bicycle in the race, having installed a single speed freewheel with an 18T (so about a 30” gear).


We would like to thank the following people for their help in this project. Shawn Buckley, Woody Flowers, Steve Loutrel, Ralph Whittemore, Ralph Bowley, Fred Anderson, and Edward Harrow.

A video by John Allen to go with this article:
Harriet's presentation on the MIT aluminum bicycle frame project at the 2016 International Cycle History Conference.

Spoke Divider

Spoke Divider

Articles by Sheldon Brown and Others

Reports of the demise of this Web site are greatly exaggerated! We at thank Harris Cyclery for its support over the years. Harris Cyclery has closed, but we keep going. Keep visiting the site for new and updated articles, and news about possible new affilations.

Copyright © 1997, 2007 Sheldon Brown

Harris Cyclery Home Page

If you would like to make a link or bookmark to this page, the URL is:

Last Updated: by Author